4.

Unit Design:
Reactor R201 - Paul O’Regan.
4.1.1 Background:

The unit operation that has been designed for the purpose of this report is Plug Flow Reactor for the
urea alcoholysis reaction of Urea and Methanol.

The Urea alcoholysis reaction occurs at 160℃ (433K), 20 bar pressure and is a known reversible
reaction. For the basis of this project, a tubular reactor has been identified as the most suitable,
followed by immediate stripping of the ammonia gas after the reaction. The reaction between urea and
methanol is a mildly endothermic reaction and for the purpose of this design no catalyst was used, due
to the immediate ‘flashing’ of the ammonia after the reaction. This removal of ammonia by ‘flashing’,
occurs in order to maintain a high equilibrium conversion. Literature suggests the use of a tubular
reactor for this process, with successful experimental data and information provided by Ranade et
al[1]. This paper suggests the use of a tubular reactor for better conversion of the urea and methanol to
dimethyl carbamate.

4.1.2 Assumptions:

It was assumed there was perfect mixing in mixing tank M100, prior to reaching the reactor.

It was assumed that mass transfer was negligible, with the reaction occurring homogeneously and no
catalyst involved.

It was assumed that all urea was completely dissolved in methanol.

It was assumed that pressure drop is negligible in the reactor.

Laminar flow was assumed in order to enhance mixing and to avoid axial dispersion.

4.1.3 Aims:

Provide a clear overview of reaction kinetics.

Use kinetics and PFR design equations to design a model for this urea alcoholysis reaction.

Calculate Q and the flow of steam required for this reaction.

Determine the material of construction for the reactor.

Evaluate heat integration opportunities.

Required Methyl carbamate production rate is 10 tonnes/hour.

4.2 Kinetics:

Literature has provided kinetics for the reaction based on experimental data. This reaction is a
reversible reaction, and the following kinetic equations were used to express the concentration of all
components in the reaction as a function of Urea.

4.2.1 Rate Law:

1
r U =k f (CU C ME− C C )
Keq MC N
C U =Cu ¿ L ( 1−X )

C Me =Cu ¿L (α −X )
¿
C MC =C N =C U , L ( X)

General PFR Design Equations:

In- Out + Generation = Accumulation

V
F A 0−F A +∫ rAdV =dN A /dt
0

At steady state:

d NA
=0
dt
Differentiating:

d FA
=r A
dV
Conversion:

F A 0 −F A
X=
FA0
Integral form:
X
dX
V =F A 0 ∫
0 −r A

4.2.2 Phase Verification:

Firstly, the Antoine equation which is a derivation of the Clausius Clapeyron equation was applied to
identify the vapour pressures of the reactants at reactor temperature of 433K. The Antoine Equation is
represented below.

B
logP (10)= A−( )
C+ T

 Temperature is expressed in Kelvin.

 A, B and C are unitless Antoine constants.
 Pressure is expressed in bar.

1. Methanol:
T = 433K

[2]

Pvap=10 ¿
Pvap=17.2 @
¯ 433 K −Methanol is∈liquid phase @ 20 .̄

2. Urea:
T = 433K

[3]

Pvap=10 ¿

Pvap=2 x 1 0−4 @433

¯ K −Urea is∈liquid phase @20 .̄

3. Methyl Carbamate:
T=433K.
T boiling point of Methyl Carbamate = 443.15K.
 Methyl Carbamate is liquid at 433K.

4. Ammonia:
T=433K

[4]

Pvap=10 ¿

P=171 @
¯ 433 K− Ammonia is∈vapour phase at 20 .̄
From the above calculations at reactor temperature, it can be concluded that the methanol, urea
and methyl carbamate are in liquid phase and the ammonia is in vapour phase at the reaction
temperature of 433K and reactor pressure of 20 bar.

4.2.3 Flow rates and Initial Concentrations:

The total flow (FT) for the R201 reactor is 10,000kg/hr. This figure was converted to different
components for the Initial Volume flow, the initial concentrations of urea and methanol and the
total volume flow.

1. Initial concentrations of Reactants:

Convert kg/hr to m3/s:

KG 1 hr 1m3
10,000 =
| |
hr 3600 s 844.6 kg
m3
=0.0033 m3 / s

Average density: 844.6 kg/m3

Density of Urea x Mass Fraction of Urea + Density of Methanol x Mass Fraction of Methanol

(1323kg/m3 x 0.1) + (791.4kg/m3 x 0.9) = 844.6 kg/m3.

Convert m3/s to L/s to calculate Vo: 3.3L/s.

0.0033 m3/s x 1000 L/m3 = 3.3 L/s

 Vo = 3.3 L/s.

Flow of Urea (mol/s): Furea (mol/s)

Total Flow Rate x Mass Fraction of Urea = Mass Flow Rate of Urea (kg/hr).

(10,000 kg/hr x 0.1) = 1000 kg/hr of Urea.

Molecular weight of Urea = 60g/mol = 0.06 Kg/mol

Kg 1hr 1 mol
Furea=1000 ∨ |
hr 3600 s 0.06 Kg |
=4.63 mol /s

Furea = 4.63 mol/s.

[A]o Urea:

[ A ] o Urea= Furea
Vo

mol
4. 63
[ A ] o Urea( mol )= s
.
L 3.3 L/ s

[ A ] o Urea=1.4 mol
L

Flow of Methanol: Fmethanol (mol/s).

Total Flow Rate x Mass Fraction of Methanol = Mass Flow Rate of Methanol (kg/hr).

(10,000 kg/hr x 0.9) = 9000 kg/hr of Urea.

Molecular weight of Methanol = 32g/mol = 0.032 Kg/mol

Kg 1 hr 1 mol
Fmethanol=9000 ∨ |
hr 3600 s 0.032 Kg |
=78.13mol /s

Fmethanol = 78.13 mol/s.

[B]o Methanol:

Fmethanol
[ B ] o Methanol=
Vo
 mol
78.13
[ B ] o Methanol ( mol )= s
.
L 3.3 L/s

[ B ] o Methanol=23.7 mol
L

4.2.4 Rate Constant:

The Arrhenius equation was used to calculate the rate constant for the reaction. Literature values
for the Pre-Exponential factor (Ao) and Activation Energy (Ea) were taken from Zhao et al [5].

[5]

The Arrhenius Equation is provided below.

− Ea
( )
RT
k = Aoex p

Ao = 1.2 x 106 L/mol.s.

Ea = 84,700 J/mol.

k = 7.3 x 10-6 L/mol.s.

4.2.5 Equilibrium Constant:

The equilibrium constant Ke was calculated using the following thermodynamic information.

Firstly, the ∆H298 and ∆S298 were calculated using the Enthalpies and Entropies of Formation for
the reactants and products at 298K. These formulas are provided below.

∆ H 298 = { ∆ H 298 products } −{ ∆ H 298 reactants } .

∆ S298 = { ∆ S 298 products }−{∆ S 298 reactants }

The values and calculations are represented in the figure below.

In order to calculate the Equilibrium constant Ke, the change in enthalpies and entropies for the
individual components had to be calculated. Once the enthalpies and entropies were calculated,
they were taken away from ∆H298 and ∆S298 to calculate ∆Hrxn and ∆Srxn. Once these values were
calculated, the below equations could be applied to calculate ∆Grxn and then finally the
equilibrium constant Ke at 433K[6].

∆ Grxn=∆ H rxn−T (433 K )(∆ S rxn)

∆ H rxn =∆ H 298 −∆ H A −∆ H B−∆ H C −∆ H D

∆ S rxn=∆ S298 −∆ S A −∆ S B −∆ S C −∆ S D

∆HA =13,535 J/mol and represents the enthalpy change of methanol at 298K to 433K (Reactor
Temperature) provided in the table below.

∆HB represents the enthalpy change of urea at 298K to 433K (Reactor Temperature) and is
calculated using the formula below.

405.8 K 433
∆ H B= ∫ Cp ( crystal ) dT + ∆ fusH ( 405.8 K )+ ∫ Cp(liquid)dT
298.15 405.8

∆fusH (405.8K) of Urea = 14790 J/mol

Specific Heat Capacity constants of Urea in crystalline and Liquid form are provided in the table
below:
Cp liquid integrated from the phase change temperature to the reaction temperature produced the
following value:

Cp crystal integrated from 298K to the phase change temperature of 405.8K produced the
following value:

∆H
B=¿14790 ( molj )+4000 ( molj )+11400 ( molj )=30,190 J / mol¿
∆HC represents the enthalpy change of Methyl Carbamate at the enthalpy of fusion 328K to 433K
(Reactor Temperature) and the enthalpy change from standard temperature 298K to the enthalpy
of fusion 328K is calculated using the formula below.

328 K 298 K
∆ HC= ∫ Cp ( liquid ) dT −∆ fusH ( 328 K ) + ∫ Cp(crystal) dT
433 K 328 K

∆fusH (328K) of Methyl Carbamate = 16700 J/mol

Specific Heat Capacity constants of Methyl carbamate in crystalline and Liquid form are provided
in the table below:
Cp crystal integrated from 328K to the phase change temperature of 298K produced the following
value:

Cp liquid integrated from the reaction temperature to the phase change temperature produced the
following value:

∆H
C=¿−16700 ( molj )+(−2570 )( molj )+(−16800 )( molj )=−36,070 J /mol ¿
∆HD represents the enthalpy change of Ammonia at the enthalpy of fusion 328K to 433K (Reactor
Temperature) and the enthalpy change from standard temperature 298K to the enthalpy of fusion
328K is calculated using the formula below.

298 K
∆ H 4= ∫ Cp ( gas ) dT
433 K

Cp gas integrated from reaction temperature 433K to standard temperature of 298K produced the
following value: ∆HD =-5170 J/mol

The table below represents the enthalpy change of methanol at 298K to 433K (Reactor
Temperature).
∆ H rxn =∆ H 298 −∆ H A −∆ H B−∆ H C −∆ H D

∆ H rxn =12,660 ( molJ )−13,535( molJ )−30,190( molJ )−(−36,070)( molJ )−(−5170)( molJ )
J
∆ H rxn =10,175( )
mol

∆SA represents the entropy change of methanol from standard temperature 298K to the reactor
temperature and is calculated using the formula below:

433 K
Cp
∆ SA= ∫ ( liquid ) dT
298 K T

J
∆ S A =36.95 /K
mol

∆SB represents the entropy change of urea in crystal form at 298K to 433K (Reactor Temperature)
and from liquid form from the phase change temperature to the reactor temperature. This is
calculated using the formula below.

405.8 K 433
Cp ( crystal ) ∆ fusH Cp
∆ S B= ∫ dT + +∫ (liquid ) dT
298.15 T ( 405.8 K ) 405.8 T

∆fusH (405.8K) of Urea = 14790 J/mol

Specific Heat Capacity constants of Urea in crystalline and Liquid form are provided in figure X
above.
 ∆S J
B=¿75.48 /K ¿
mol

∆SC represents the enthalpy change of Methyl Carbamate at the enthalpy of fusion 328K to 433K
(Reactor Temperature) and the enthalpy change from standard temperature 298K to the enthalpy
of fusion 328K is calculated using the formula below.

328 K 298 K
Cp ∆ fusH Cp
∆ SC = ∫ ( liquid ) dT − +∫ (crystal )dT
433 K T ( 328 K ) 328 K T

∆fusH (328K) of Methyl Carbamate = 16700 J/mol

Specific Heat Capacity constants of Methyl carbamate in crystalline and Liquid form are provided
in Figure X above:

∆S J
C =¿−102.67 /K¿
mol

∆SD represents the entropy change of Ammonia at the enthalpy of fusion 328K to 433K (Reactor
Temperature) and the entropy change from standard temperature 298K to the enthalpy of fusion
328K is calculated using the formula below.

298 K
Cp ( gas )
∆ S 4= ∫ dT
433 K T

Cp/T integrated from reaction temperature 433K to standard temperature of 298K produced the
following value: ∆SD =-14.5 J/mol/K

∆S J
D=¿−14.5 /K¿
mol

Entropy of reaction at 433K was calculated by inputting the calculated individual and entropies into
the below equations:

∆ S rxn=∆ S298 −∆ S A −∆ S B −∆ S C −∆ S D

J J J J J
∆ S rxn=38.8
mol
K ( ) ( ) ( ) ( )
−36.95
mol
K
−75.48
mol
K
+102.67
mol
K
+14.5(
mol
K
)

J
∆ S rxn=43.54 /K
mol
The Gibbs Free Energy was then calculated at reactor temperature of 433K using the following
formula:

∆ Grxn=∆ H rxn−T (433 K )(∆ S rxn)

J
J mol
∆ Grxn=10,175
mol ( )
−43.54 (
K
)(433 K)

∆ Grxn=−8677.8 J /mol

The equilibrium constant was then calculated using the Gibbs Free Energy at reactor temperature
using the following formula:

−∆ G
( )
RT
Ke=exp ⁡

J
−−8677.8
mol
( )
8.314 J
mol
∗( 433 K )
K
Ke=exp ⁡

Ke=11.14

4.2.6 Fractional Conversion:

The fractional conversion for the reaction was then calculated to be 92% conversion, using the
below equation:

Ke
Xe=
1+ Ke

11.14
Xe=
1+ 11.14

Xe=0.92

Xe=92 % conversion

4.3 Reactor Volume:

The reactor volume was calculated using the Ordinary Differential Equation Solver Polymath.
The above calculations were used and entered into the software as inputs, in order to calculate the
volume of the reactor with respect to conversion. The figure below represents the inputs and units
used in Polymath to calculate the reactor volume.
The reactor volume was calculated to be 36,000 L or 36 m3. The calculated values are provided in
the Polymath produced report below.
The graph below represents how conversion increases as reactor volume increases.
The graph below represents the formation of Methyl Carbamate and Ammonia as the reaction
progresses.

4.4 Residence Time:

The residence time tr was calculated using the following formula, as the reaction took place in the
liquid phase.

V (L)
tr (s)=
L
vo( )
s

36,000 L
tr (s)=
3.3 L
s

tr=10,909 s

10,909 s
tr=
3600 s

tr=3 hours

This result states the reaction requires 3 hours to achieve 92% conversion of Urea to Methyl
Carbamate and Ammonia.
4.5 Reactor Heat Properties:

Pressurized steam at 20 bar was selected as the heating utility to heat the reactor from 37.5℃ to
the reactor temperature of 160℃. The overall heat transfer coefficient U for the reactor was
selected to be 1100 W/m2/K as there is steam outside the tubular reactor and liquid inside[7].

Tube diameter was selected to be 0.02m and tube length was selected to be 6m for the reactor.
The surface area of one tube was calculated using the formula for the surface area of a cylinder
provided below.

Surface Area of One Tube=2 πrh

SA=2∗( 3.14 )∗( 0.02 m )∗6 m

SA=0.3768(m2)

The Volume of one tube was then similarly calculated using the formula for the volume of a
cylinder:

Volume=π ( r 2 ) h

Volume=3.14∗( 0.01 m 2)∗6 m

Volume of 1 tube=0.0018 ( m 3 )

The total number of tubes in the reactor was then calculated to be 15,960 tubes using the Total
reactor volume and the volume of one tube.
 Reactor Volume(m 3 ) 36,000
Total number of tubes= = =15,960tubes
Volume of 1tube(m3 ) 0.0018

Individual Tube specifications:

The overall heat duty for the reactor was calculated using the formula:

Q= { Δ H } ∈− { Δ H } out

Table for ∆H for the inlet stream and subsequent conversions to (J/s).

Table for ∆H for the outlet stream and subsequent conversions to (J/s).

J
Q=−23,242.8( )−785.9(J /mol)
mol

Q=−24,028.7 ( Js )=Overall Heat Duty of Reactor

The general equation for heat transfer was used to calculate the estimated surface area of the reactor.

Q=UA Δ T lm

Where Q is the heat duty, U is the overall heat transfer co-efficient, A is the surface area and ΔTlm is
the mean temperature difference.
∆Tlm for the reactor was calculated using the below equation:

Δ Tlm=¿ ¿

THot In – Hot fluid temperature in. (310.7K)

TCold in – Cold fluid temperature in.(443K)

THot Out – Hot fluid temperature out. (433K)

TCold Out – Cold fluid temperature out. (423K)

Δ T lm=42.3 K

The estimated heat transfer area was then calculated using the general heat transfer equation above.

An estimated tube surface area was then calculated by multiplying the total number of tubes (15,960)
by the individual tube surface area (0.378m2), giving a value of 6014m2. Using this value an estimated
heat transfer coefficient value could be calculated, using Q, A and ∆Tlm.

24,028.7 ( Js )=U∗( 6014 m )∗( 42.3 K )

2

W
U =0.09
m 2/K

4.5.1 Reactor and Tube parameters:

These parameters were calculated in the below tables and were heavily reliant on outlet mass fraction:

These values were used to calculate further specifications for the reactor provided in the below
equations:

Number of tubes 15,960

Nttp= = =15,960tubes
Number of Passes 1
 Tube side Mass Flow
Gm=
Nttp∗¿ ¿

0.55
Gm 673.22 kg
Tube side velocity= = =0.0008
pf m3

The Reynolds number and Prandtl number were calculated for the fluid using the following equations:

Pr=μCp/k

1
pf 4 3
−5
k =3.56∗10 ∗Cp∗(
Mw ( )
)

1.93∗10−4∗3.67 −3
Pr= =3.01∗10
0.24

ρ∗v∗di
ℜ=
μ

673∗0.0008∗0.02
ℜ= =57.3
1.94∗10−4

As the Reynolds number was below 5000, this verifies that fluid is in laminar flow for the reaction
which was one of the prior assumptions. This helps avoid axial dispersions in the fluid. The heat
transfer coefficient (hi) was then calculated for Laminar flow using the equation below:

[8]

4.5.2 Shell parameters:

The mass of steam required was calculated in (kg/s) using another standard equation for heat transfer:

Q ( kjs )=mCp ΔT
mass of steam(m)=0.4 kg/s

An outer diameter of 0.002m was selected for tube thickness, making the outer diameter of the tube to
be 0.022m.

The following formulas were then used for the calculations involving the shell side of the reactor.

Tube pitch( pt)=1.25∗do=0.0275 m

Nt 1n
Bundle diameter ( BD )=do ( )
K1
=3.5 m

Where K1 and n1 were obtained for 1 pass from the table below:
 [8]

Assumed shell diameter = bundle diameter, shell diameter (Ds) is 3.5m.

Baffle spacing(BS)=0.4∗(Ds)=1.4 m

The area for cross flow (As) and the shell side mass velocity (Gs) were calculated using the below
equations. Other parameters such as shell equivalent diameter were also calculated and provided in
the figure below.

[8]

The shell side Reynold’s number and Prandtl number were then calculated using the formulas below:

Gsde 0.4∗0.022
ℜ= = =24.6
μ 3.5∗10−4

Pr=μCp/k

3.5∗10−4∗3.02
Pr= =1.19
0.0009

jh was obtained using the Reynolds number and the chart below to produce a value of 0.1.
[8]

The Nu number was then calculated using the formula:

1 0.14
μ
Nu= jhReP r 3∗ ( )
μw
=2.6

The heat transfer coefficient for the shell side was then calculated using the formula for Laminar flow
again:

The heat transfer coefficient was calculated to be 0.04, and the parameters are represented in the
figure below:

The Overall heat transfer coefficient Uo was calculated using the formula below:
 1 1 do
do∗10−3∗ln ( dodi )
1
(
= +0.00035
Uo hi
+ )( ) + + 0.0002
di 2∗k (thermal conductivity of material) hs

This is represented in the figure below:

The values for the heat transfer coefficient do not match up currently, however trial and error is being
continued using the above process in order to get the U values to within 10% of one another.

4.6 References:

1. Ranade, V.V., et al., Configuration and its use in process for synthesis of alkyl carbamates
from alkyl alcohol and urea in a tubular reactor. 2019, Google Patents.
2. NIST.gov, Methyl alcohol. 2018.
3. Lundström, A., et al., Modelling of urea gas phase thermolysis and theoretical details on urea
evaporation. Proceedings of the Institution of Mechanical Engineers, Part D: Journal of
Automobile Engineering, 2011. 225(10): p. 1392-1398.
4. NIST.gov, Ammonia. 2018.
5. Zhao, W.-b., et al., Macro kinetics for synthesis of dimethyl carbonate from urea and
methanol on Zn-containing catalyst. Journal of Central South University, 2012. 19(1): p. 85-
92.
6. Zeng, Z.-X., et al., Heat capacity, enthalpy of formation, and entropy of methyl carbamate.
Industrial & engineering chemistry research, 2010. 49(12): p. 5543-5548.
7. engineersedge.com, Overall Heat Transfer Coefficients Table Charts and Equations. 2021.
8. Shawabkeh, R.A., Steps for Designing Heat Exchanger. 2015.
4.2 Design Specifications Reactor R302

In reactor R302, methyl carbamate, MC and methanol, MeOH react to form DMC and
ammonia, NH3
M C (l )+ MeO H (l)=DM C (l) + N H 3(l )

The simultaneous stripping of ammonia gas and DMC from the liquid phase is achieved by
dispersing superheated methanol vapour (stripping gas) which is fed to the bottom of the
reactor The liquid reactant feed contains a small amount of catalyst - samarium nitrate which
has been previously liquified. The homogenous catalysis increases DMC selectivity – 65.8%
and methyl carbamate conversion 43.1% and reduces the formation of the side product MMC
– 4%. The overhead vapour product stream contains ammonia, the DMC:Methanol azeotrope
and some impurities. The bottom liquid product stream contains MC, catalyst, and some
impurities. The MC in this stream is purified and recycled back as an additional feed inlet
stream of the reactor. Th operating conditions of the reactor is 180⁰C and at 20 bar.

Inlet Streams
Stream Phase Key Composition Flow Rate kg/hour

325 Vapour MeOH 21,000

323 Liquid Recycled MC 9250
302 Liquid Methanol /Catalyst 9045.68

Outlet Streams

303 Vapour NH3, DMC: Methanol 29,4717

315 Liquid MC, Impurities 9823.92

Table 1 ; Inlet and Outlet Streams information R302

4.3 Reactor Conditions

The reactor conditions are estimated at 180⁰C and 20 bar. The Antoine Equation is used to
confirm that at these conditions the reactants, methyl carbamate and methanol are in liquid
phase, and that the DMC and NH3 are formed in the liquid phase of methanol.

log P10= A− ( C+B T )

where P is reaction pressure (bar) and A, B and C are Antoine constants, and T is temperature
(K)
Temperature = 453K, P = 20 bar
Methanol
Antoine Constants
A 5.31
B 1676.6
C -21.728

1676.6
5.31−(−21.73+ 453 )
Pvap =10 =26.45 ¯¿

Thus, methanol is in liquid phase at 453K and 20bar

Antoine constants for methyl carbamate, ammonia and DMC are not found at 453K. The
Clausius Clapeyron equation is thus used to determine the vapour pressure.
P2 ∆ H vap 1 1
ln = ( − )
P1 R T2 T1

where T1 is the boiling temperature (K), T2 is the reaction temperature (K) R is universal gas
constant, P1 is atmospheric pressure (atm) P2 is the reaction pressure (atm) Hvap is heat of
vaporisation (J/mol)
Methyl Carbamate
P2 ?
P1 1
T2 453
T1 450
ΔHvap ?

Ammonia
P2 ?
P1 1
T2 453
T1 239.6
ΔHvap 23350

P2 −23350 1 1
ln = ( − )
1 8.314 453 239.6
Pvap =1.05 ¯¿

Ammonia is vapour at 20 bar and 453K

DMC
P2 ?
P1 1
 T2 453
T1 363
ΔHvap 3690000

P2 −369000 1 1
ln = ( − )
1 8.314 453 363

Pvap =3.55 ×1 010 ¯¿

Thus, DMC is formed in the liquid phase.

4.3 Determining Fractional Conversion

MC + MeOH=DMC + N H 3

Equilibrium Constant Ke :

−∆ G
K e =exp
RT

where R = universal gas constant 8.314 J/mol K

Where ΔG = Gibbs Free Energy

∆ G=∆ H−T ∆ S
Where ΔH = enthalpy, ΔS = entropy

From Energy Balance, enthalpy of reaction is calculated from the predetermined enthalpy of
reactants and products at 453K

Reactants Products
Enthalpy J/mol Enthalpy J/mol
MC - Liquid 1.43x104 DMC – Liquid 1.74x104
MeOH - Liquid 7.74x103 NH3 – Liquid 5.9x103
*wrong enthalpy values

∆ H= ∑ ∆H− ∑ ∆ H=1260 J /mol

products reactants

Similarly, for entropy ΔS = 98.14J/mol K

 Reactants Products
Entropy J/mol K Entropy J/mol K
MC - Liquid 77.1 DMC – Vapour 109.66
MeOH - Liquid 127.19 NH3 – Vapour 192.77

Thus giving:

∆ G=∆ H−T ∆ S=1260−( 453 × 98.14 )=−4.3 ×10 4 J /mol

−∆ G 4.3 ×10 4 J /mol 4

K e =exp =exp =9.5 × 10
RT 8.315 × 453 K

wrong

Temperature Rate Constant Activation Energy, Ea Preexponential Factor,

A
-1 -1
K mol . L . s kJ mol mol . L . s-1
453 1.65x10-6 75.3 8.12x102
Table 1 :Kinetics of DMC [5]

Reaction rate constant, k

−Ea
k = Aexp
RT

−75300
k =8.12 ×102 ×exp =1.68× 10−6 mol L s−1
8.314 × 453

Calculating Xe:

1. Volume flow rate of Stream 302 (Liquid Inlet Stream)

9045.68 kg 1 hour 1 m3 6.19x10-3 m3

hour 3600 seconds *406.059kg s

*Average density

MeOH = density of methanol x mass fraction = 792kg/m3 x 0.87 = 689.04kg/m3

MC = 1136kg/m3 x 0.1 = 118.68kg/m3

Average density = 406.059kg/m3

2. Molar Flow rate of MC in 302:

9045.68 kg 0.10 1 hour 1 mol 3.5 mol

hour 3600 seconds 0.075kg s

3. Concentration of MC in 302

3.5 mol /s
[ MC ] = =0.565 mol /L
6.19 L/s

4. Similarly, the concentration of methanol in 302

[ MeOH ] = 68.66 =11.09 mol /L

6.19

5. Volume flow rate of stream 323 (Recycled MC stream)

9250 kg 1 hour 1 m3 2.26x10-3 m3

hour 3600 seconds 1136kg s

6. Molar Flow rate of MC in 323:

9250 kg 1 hour 1 mol 34.26 mol

hour 3600 seconds 0.075kg s

7. Concentration of MC in 323

[ MC ] = 32.26 mol /s =14.27 mol /L

2.26 L/s

8. Total Concentration of MC

14.27 M +0.565 M =14.835 M

5. Similarly, the concentration of DMC (1MC:1DMC)

[ DMC ] =14.835 M

5. Similarly, the concentration of NH3 (1MC :1NH3)

[ N H 3 ]=14.835 M

M C (l )+ MeO H (l)=DM C (l) + N H 3(l )

C2H5NO2 CH3OH C3H6O3 NH3

N0 (mol/L) 14.835 11.09 14.835 14.835

@equlibrium A0 (1-Xe) B0 -A0Xe C0 + A0Xe D 0 + A0Xe

(mol/L) 14.835 (1-Xe) 11.09-14.835Xe 14.435+14.435X 14.435+14.435Xe
e

C c D d ( 14.435+14.435 Xe ) ×(14.435+ 14.435 Xe) 4

K e= a b
= =9.5× 10
A B (14.435−14.435 Xe) ×(11.09−14.435 Xe)

3.03 Xe2 −62.6 Xe=−59.5

Xe=1

Wrong as thermodynamics properties values In wrong phase

4.4 Desorption Column Equations

The design of the stripping column focuses on the desorption rate (forward reaction)
of ammonia from the liquid methanol phase to the superheated methanol phase by mass
transfer. The reaction operates with fast kinetics and mass transfer.
 Figure 1; Desorption Column Basis

L’ = Molar Flow rate of solute free absorbent

V’ = Molar Flow of solute free gas

X = Mole Ratio of solute to solute free absorbent in the liquid

Y = Mole Ratio of solute to solute free gas in the vapour

The values of the liquid (L’) and gas (V’) flow rates remain constant throughout the column.

Stream 325 contains the, stripping gas superheated methanol vapour (0.99) and has a
mass flow rate of 21,000kh/hr. Thus, the stripping gas has an individual mass flow rate of
20966kg/hr, which is converted to kmol/hr :

20790 kg 1 mol 1 kmol 654.38 kmol

hour 0.03204kg MeOH 1000mol hour

Superheated methanol stream contains no ammonia so inert gas molar flow rate:
V’ = 654.38/hour

The molar ratio of NH3 in the inlet gas stream is

Y0 = 0

Liquid flow rate contains 94% of methanol results in 8502.94kg/hour mass flow rate of
methanol which is converted to kmol/hr

8502.9 kg 1 mol 1 kmol 265.38 kmol

hour 0.03204kg MeOH 1000mol hour
The mole fraction of ammonia in the inlet liquid stream :
- 32.5% MC converted to Ammonia
- MC present in two feed streams:
Stream Mole Fraction of MC Conversion to Ammonia

302 0.048 0.0156

323 0.99 0.3223
0.3379

Inert Liquid flow rate = 265.38 x (1-0.3379) = 175.7kmol/hr

L’ = 175.7kmol/hr

The molar ratio of NH3 in the liquid phase:

0.3379
X 0= =0.6621
1−0.3379

The mass balance for ammonia:

X N +1 L ’+Y 0 V ’=X 1 L’ +Y N V ’

0.6621 ×175.7+0=0+Y N ×654.38

Y N =0.177

L’ 175.7kmol/hour
V’ 654.38kmol/hour
Y0 0
XN+1 0.6621
X1 0
YN 0.177

X n+1 > X1 , YN > Y0

The flowrate of NH3 to be removed,

654.38−175.7=478.68 kmol/hr
4.5 Type of Column Reactors
The main types of desorption columns exist as packed or trayed column reactors. Tray
columns are equilibrium based, whereas packed columns are rate based. Other types of
columns include bubble column reactors that can facilitate crossflow stripping which is more
effective than counter-current flow in packed and tray column reactors.

4.5.2 Tray Columns

In tray column design, gas enters the bottom of the column and rises. As it rises (due
to convection) in comes in contact with the liquid stream which is flowing downwards (due
to gravity) and usually counter-currently. In tray column design, weeping, entrainment, and
downcomer flooding are some of the design parameters that must be considered. The flow
regimes of the gas-liquid flow must too be considered.

The internals of a tray column including tray diameter, height between trays pressure
drop and efficiency must be determined. Common tray designs include bubble cap trays,
valve trays and sieve trays.

4.5.3 Determining the Number of Stage Required (Tray Column)

From the mass balances the operating line can be determined with a slope of L’/V’:

L' L'
Y n= X n+1 ( )+Y 0 – X 1 ( )
V' V'

L' 175.7
= =0.268
V ' 654.38
Y n= X n+1 (0.268)+0

The equilibrium line for the NH3 at stage n :

y n Y n /( 1+Y n)
K n= =
x n X n /( 1+ X n )

Kremser Method Analysis to determine the number of stages can be applied to this
reactor as there is a dilute concentration of solute, and one or more components are being
stripped – NH3 and DMC. The fraction of solute, i, stripped can be denoted as :

S Ni +1−Si
N +1
S i −1
where Si is the stripping factor;

KV '
Si=
L'

K equilibrium constant mainly depends on temperature and pressure, V’ is the molar flow
rate of solute free gas, and L’ is the molar flow rate of solute free liquid.

P¿i ¯
K= =21.3 ¯ ¿ ¿
P 20 ¿ =1.07 ¿

Where Pi* = partial pressure

P
P¿i = =20 ¯¿ =21.3 ¿
Xa 0.94

Si=1.07 x 0.268=0.28

Very small stripping factor; optimum stripping factor 1.4.

To determine the equilibrium line, henrys law constant, H is used to determine the slope, m.
The equilibrium can be straight or curved.

y=mx+c

Ci 0.032
m=H = ¿= =1.50 ×1 0−3 mol / L ¯¿
Pi 21.3

To determine the number of stags required using fractional conversion, Xe of ammonia =

0.325 (Literature)

Total moles of Methyl Carbamate, MC enter reactor:

Stream Moles of MC
302 0.139

323 1.331

Total In = 1.47
Number of moles of Ammonia to be removed; 0.191 moles

Stages #moles of MC Xe #moles converted to

NH3
1
1.47 0.325 0.477737
2
0.992223 0.325 0.322472
3
0.66975 0.325 0.217669
4
0.452081 0.325 0.146926

To achieve 0.191 moles removed required 4 stages – (removal of 0.147 moles)

Figure 2 : Screenshot of ASPEN plus results using 4 stages of sieve trays for the required separation

4.5.4 Packed Bed Column Reactor – Mass Transfer Based

A packed column reactor operates with a continuous gas-liquid contact, unlike the
stage wise approach in a tray column. The gas flows upwards and the liquid flows
downwards similarly to a plate column. Optimum gas and liquid distribution is important for
packed column design.
 The packing must provide a large surface area for efficient mass transfer between the
liquid and the gas, have an open structure to ensure low resistance to gas flow, promote
uniform liquid distribution on the packing surface and promote uniform vapour gas flow
across the column cross section.

The two main types of packing are random packing and structured packing. Random
packing are placed in the column to take up random arrangement - such as Raschig rings and
Cascade rings, berl saddles and Intralox saddlers. Structured packing has regular geometry
usually made from wire gauze and mesh.

Packed column internals focus on the mass transfer rate which depend on the mass
transfer coefficient kG (gas) and kL (liquid). Liquid hold up, diameter, flooding, pressure drop
and mass transfer efficient must be determined.

4.5.5 Mass Transfer Calculations

Volume of Superheated Methanol Vapour
nRT 3.120 ×8.314 ×453
V= = =0.00588 m3
P 2 ×1 0 6

Solubility of ammonia in SH MeOH vapour

0.191moles N H 3
=0.032
5.88 L SHMeOH

Diffusivity Coefficient, DAB using Fullers Equation;

D AB=
10−7 T 1.75
(√ M1 + M1 )
A B
1 1 2
P [ (∑ V A
3
) (∑ V B ) 3
]
Methanol: CH3OH
⅀VA= 16.5 + (4 x1.98) +5.48 = 29.9g/mol
Ammonia: NH3
⅀VB= 5.69 + (3x1.98) = 11.63g/mol
MA = 32.04g/mol, MB = 17.3g/mol
P = 20 bar
Temperature = 453K

1 1
D AB=
10−7 4531.75 (√ 32.04 +
17.3 )
=2.34 × 10 −6
m2 / s
1 1 2
20 [ ( 29.9 ) ( ∑ 11.63 ) ]
3 3

4.5.6 Column Diameter

The capacity of a column is determined by the cross-sectional area. The capacity of
the column should be designed to facilitate the most sufficient economical pressure drop that
ensures thorough liquid and gas distribution.

Gas flow rate (superheated methanol);

5.83 kg /s
=0.182kmol /s
32.04 g /mol

Liquid Flow rate (methanol)

9045.68 x 0.94 2.36 kg 2.36

= = =0.074 kmol /s
3600 s 32.04

Type of Packing: 33mm ceramic Intalox Saddles

Gas Density

M w × moles 0.03204 kg /mol ×3.12 moles 3

= 3
=17 kg/m
Volume 0.0059 m

Liquid density; 406.06kg/m3

Liquid Viscosity = 1.4 x 10-5 Pa.s

The Flow Factor, FLV

L¿W ρ V 0.074 17
F LV = ¿
VW √ =
√
ρ L 0.182 406.06
=0.009
 Figure 2: Pressure Drop Correlation to calculate K4

Flooding Factor K4

¿ 2 μ L 0.1
13.19 ( V W ) F p ( )
ρL
K4= =0.7
ρV ( ρ L −ρV )

K4 is obtained graphical from Figure using 20mm of water/metre of packed height.

Rearranging;

1/ 2 1 /2
K 4 ρV (ρ L −ρV ) 0.7 ×17 ×( 406.06−17)

[ ][ ]
¿ 2
V =
W = =3.4 kg /m s
μ L 0.1 1.4 ×10−5 0.1
13.1 F p ( ) 13.1×170 ×( )
ρL 406.06

Required Column Area

5.83 kg /s
2
=1.7 m 2
3.4 kg /m s

Diameter
 4
√ π
× 1.7=1.47 m

4.5.7 Column Height

The height of the column is determined by calculating the height and number of
transfer units, HTU and NTU respectively.

The concentration of ammonia is less than 10%, and V’ and L’ are constant
throughout the column. The height of the packing can be therefore be calculated by;

Z=HTU × NTU

where HTU is the height of the overall liquid-phase transfer unit, and HTU is the number of
overall liquid-phase transfer units. HTU and NTU are expressed as followed;

x1
L' dx
HTU = NTU =∫
K La Ct x2
x e −x

where L’ = molar gas flow-rate per unit cross sectional area, k L = liquid-phase mass transfer
coefficient, a = interfacial surface area per unit volume, Ct = total molar concentration x1 and
x2 = the mole fractions of solute in the liquid at the bottom and top of column respectively , x e
mole fraction of liquid in equilibrium with gas phase

The liquid- phase mass transfer coefficient kL must be determined to calculate the
heights of transfer units HTU. An empirical expression of kLa (units s-1) ;

k L a=78.8 D0.5
AB ( F+ 0.425)

where kL is the mass transfer coefficient, a is the interfacial surface area per unit volume,
where DAB= diffusivity coefficient cm2/s, F (F-Factor) = Uvrv0.5

Superficial vapour velocity

Q
U v=
A
where Q is the volumetric flow rate of superheated methanol vapour m3/s and A = cross
sectional area m2
Mass flow rate of the superheated methanol is 5.83kg/s
Volumetric flow rate of superheated methanol

m × R ×T 5.83 ×8.314 × 453

Q= = =0.343 m3 / s
Mw× P 0.032 ×2 ×10 6

Q 0.343
U v= = =0.202m/s
A 1.7
F-Factor
kg
F=U v ρv =0.202× 170.5 =0.83(m/s )(
0.5 )
m3
Liquid-phase mass transfer/ interfacial area kLa

0.5
k L a=78.8 D 0.5 −6
AB ( F +0.425 )=78.8 × ( 2.34 ×1 0 ) ( 0.83+ 0.425 )

¿ 0.15 s−1

Overall liquid-phase mass transfer/interfacial area coefficient, KLa

1 1 1
= +
K L a mk G a k L a

Where m is henrys constant of the system (1.50x10-3) and kG is the gas-phase mass transfer
coefficient which can be expressed as;

1.030 D0.5 2
v ( f −0.842 f )
k g a= 0.5
( hl )

175.7
HTU = =¿
Kl ×12.67
 x 2−x 1
NTU =
( x−x i )logmean

(x ¿ ¿ 1¿ −x 1)−( x ¿2−x 2)
( x−x i )logmean = ln ¿ ¿
¿

Where x1*=mx1 , x2*=mx2

Mole fraction x m slope mx

x1* 0 1.5x10-3 0
x2* 0.6621 1.5x10-3 9.9315x10-4

The mass transfer flux, NA:

c D AB 1− X A 2
N A= ln
∆z 1− X A 1

Δz = 2.4384m
XA2 = 0
XA1 = 0.032721
Concentration of ammonia = 0.191/0.360L = 0.53M

0.53× 2.34 ×1 0−6 1−0 2

N A= ln =kmol m / s
z 1−0.033

N A AdZa=k L a( C Ai−C L )

Pressure Drop

∆ P=0.115 F 0.7 0.7 −1

p =0.115 × 17 0 =4.18 m
4.1 H301 heat exchanger design.

A shell and tube heat exchanger was selected to heat methanol from 20C to 180C.
Condensing steam was selected as the heating medium on the shell side. Overall heat transfer
co-efficient was selected as 800 W/m2K this Value was selected using Figure 4.1 below. In
order to ensure safe operation of the heat exchanger, BS3274 and BS3606 were used in
determining the standards of materials for use in this heat exchanger. [9]

Figure 4.10 Overall Coefficients

The length of the Heat exchanger was selected to be 6.1m, using a longer length of results in
a smaller shell diameter. The smaller shell diameter reduces the cost of the heat
exchanger[10]

The general equation for heat transfer was used to calculate the overall surface area of the
heat exchanger.

Q=UAΔTm Eqn4.101

Where Q is the heat transfer, U is the overall heat transfer co efficient, A is the surface area
and ΔTm is the mean temperature difference.

ΔTm was calculated using ΔTlm

(T 1−t 2)−(T 2−t 1)

ΔT lm= Eqn 4.102
ln ⁡¿ ¿

T1= Hot fluid inlet temperature

T2= Hot fluid outlet temperature

t1= Cold Fluid inlet temperature

t2= Cold fluid outlet temperature

ΔTm= ΔTlm *Ft Eqn 4.103

Where Ft is the temperature correction factor. Correction factor is determined using figure
4.11. Where R and S are determined using:
(t 2−t 1 )
R= (T 1−T 2) Eqn 4.104 S= Eqn 4.105
¿¿ (T 1−t 1 )
Figure 4.11 Temperature correction factor: one shell; Two or even more tube passes.

Correction factor is 1 in this instance.

Bundle diameter was calculated using

Nt 1/n1
Db= Do( ) Eqn 4.106
K1
K1 and n1 are determined by the number of passes and the pitch.
Figure 4.11 was used to determine the inside shell diameter from the bundle diameter. A split
ring floating head was used for this heat exchanger. The shell diameter Ds was determined
from the bundle diameter and BDC graph. Baffle spacing is normally 0.3-0.5 depending on
the pressure drop in the shell side. [9]

Figure4.11 BDC
Several other equations as outlined below were used to determine the pressure drop in the
shell. [11]

Ds= Db +BDC Eqn4.107

Bs = 0.4Ds Eqn4.108

( pt−do ) Ds Bs
As= Eqn4.109
Pt

Shell side mass flow rate

Gs= Eqn4.111
As

Pt 2−πDo 2
De=
4∗ ( 4 ) Eqn4.112
πDo
 GsDe
Re= Eqn4.113
μ

μ Cp
Pr= Eqn4.114
k

h iDe μ 0.14
Nu= = jhRePr1/3( ¿ Eqn4.115
Kf μw
Where jh is obtained from figure 4.12 and Kf was calculated using figure 4.13.

Figure 4.12 Baffle cut chart.

 Figure 4.13 Thermal conductivity of Steam

Finally the pressure drop in the shell was calculated using.

Ds L ρu 2 -0.14
ΔP = 8jf ¿) Eqn4.116
De lb 2

Jf was estimated using Figure 4.14 with 35% baffle spacing.

 Figure 4.14 Shell side friction factor, segmental baffles.
For tube side pressure drop required the implementation of numerous equations detailed
below.
Ntpp = Nt/ number of passes. Eqn4.117

Tube side mass flow

Gm = Eqn4.118
Ntpp∗π Di2 /4

Tube side velocity = Gm/ρ Eqn4.119

Re and Pr were calculated using
μCp ρDiu
Pr= Eqn4.119 Re= Eqn4.120
k μ

The heat transfer co efficient for turbulent flow was calculated from

Kf 0.8 0.33 μ 0.14

0.023 Re Pr ( ) Eqn4.121
Di μw

The overall heat transfer coefficient and tube side pressure drop were calculated.
 ρu 2
ΔP=¿)-m]) Eqn 4.122
2
do
d o x ln ⁡( )
1 1 1 di d o 1 d o 1 Eqn 4.123
= + + + x + x
U ho h od 2 kw d i hid d i hi

hod outside fouling factor, hid inside dirt coefficient, kw thermal conductivity of the tube
material.
After several alterations to the length of tube, tube diameter, number of passes and inlet
temperature the overall heat transfer co-efficient balanced.
Table 4.10 design specification sheet for shell and tube heat exchanger

Shell side Tube Side

Fluid Saturated Steam Methanol
Temperature in (K) 472 293
Temperature out (K) 472 458
Pressure Drop (KPa) 25 8700
Construction material Steel alloy Stainless steel Sch 40
Area (m2) 30.8
Length (m) 6.1
Number of passes 2
Number of Tubes 95
Tube Diameter OD (mm) 0.019
Tube thickness (mm) 2
Shell Diameter (mm) 384
Shell Thickness (mm) 3.2
Baffle Spacing (mm) 153
Pitch Square 1.25Do
Flow type Counter current
Volumetric flow of steam 0.104
(m3/s)

4.2 D304 Distillation column design.

The preliminary distillation column design was undertaken using the McCabe Thiele method.
Vapor liquid equilibrium curve was determined using Ammonia and DME as light and heavy
key components respectfully. It is assumed all of the CO2 as a more volatile light non key
component will be removed in the distillate, therefore to mole fraction of the light non-key
component is included with the light key. A pseudo binary system is established, the vapor
pressure of the DME and Ammonia were calculated using Eqn 4.201. Raoult’s Law and
Daltons law were used to calculate the VLE data see appendix 4.2 for excel calculations.
 C2
ln ⁡( P)=C 1+ +C 3∗lnT +C 4 T C 5 Eqn 4.201[12]
T
where P is the pressure in Pa. and C1-C4 are coefficients.
Table 4.201 constants for Vapor pressure calculation for ammonia.
Tmin Tmax
C1 C2 C3 C4 C5 P (Pa) (K) (K)
Ammoni
a 90.493 -4669.7 -11.607 1.72E-02 1 431202.3 195 405
Table 4.201 vapor pressure of ammonia.
−4669.7
ln ( P )=90.493+ −11.607∗ln 273+ 1.72 x 10−2 273 1
273
−4669.7 −2 1

P = e 90.493+ 273
−11.607∗ln273+1.72 x10 273

P = 431202.3 Pa
Calculations were repeated for DME. The total column pressure was set to 303 kPa, from this
the column temperature at various vapor liquid compositions was determined using Excel
solver. Figure 4.2 shows the VLE curve for Ammonia and DME at 303kPa.

Figure 4.20 VLE graph.

Calculating flow rates

A mass balance for the column can be given by:
F =D +W Eqn 4.202
A component balance can then be written as:
 Fxf = Dxd + Wxw Eqn 4.203
Aspen HYSYS was used to determine the steady state flow rates at the given process conditions.
Stream 310 311 314
Ammonia (Kg/h) 467.9 0.17 467.8
DME (Kg/h) 10.4 9.8 0.6
CO2 (kg/h) 61.8 0 61.8
Table 4.202 mass flow of components
Calculation of Q-Line
The q-line is used to determine the upper and lower operation lines, and also the minimum reflux
ratio. To calculate the q line the feed properties, need to be calculated.
λ+ H fs + H f
q= Eqn 4.204
λ
λ is the molar latent heat of the vapor, H fs is the molar enthalpy of the feed, H f is the molar enthalpy
of the feed at its boiling point. The calculated values are based of information obtained from
Literature [13].

Feed Mol Fraction H fs (kJ/mol) H f (kJ/mol) λ (kJ/mol)

(Kmol/h)
Ammonia 27.48 0.944 6.18 3.36 20.98
DME 0.227 0.0078 24.1 21.4 20.58
Low Boiling 1.403 0.0482 -3.65 -10.3 10.49
impurities
Total 29.11 1 26.6 14.5 52.05
Table 4.203 latent heats
q can now be calculated using eqn 4.204.
52.05+26.6−14.5
q= = 1.23
52.05
q
q line slope can be determined using
q−1

q xf
The q line can be determined using the equation y= x− . This gives an expression for the
q−1 q−1
q line of y= 5.35x-4.1

Determining the minimum reflux

The upper operating line can be found from xd,yd and the point where the q line intercepts
the equilibrium line (xg,yg). From a similar basis the lower operating line can be found from
xw,yw and xg,yg. The minimum reflex ratio can then be calculated using the slope of the
upper operating line from the equation.
Rm y d− y g
=slope= Eqn 4.205
R m−1 x d−x g
 Rm =−0.17

The increasing the reflux ratio will reduce the number of plate required, however it will
increase the running costs as more vapor and liquid will need to be condensed and reboiled.
The figure below demonstrates the effect of reflux ratio on cost.

Figure 4.21 effect of reflux ratio on cost

Determining the number of Plates:

The reflux ratio was set to 3 and the McCabe Thiele method was initially used to determine
the number of theoretical plates needed. As can be seen from figure 4.4 it was not possible to
accurately assess the number of plates using this method, as a result Aspen HYSYS was used
to determine the theoretical and actual number of plates required. Antoine fluid package was
used for the simulation.
 Figure 4.22 McCabe Thiele method to determine the theoretical number of plates
The initial feed composition were determined using mass balances. Top and bottom products
were preset based on the quality of product needed for resale. Table 4.204 shows the
compositions of each stream. On setting the required distillate and bottom product
compositions it was established a minimum reflux ratio of 1.52 was needed to achieve the
desired separation.

Figure 4.23 Aspen HYSYS column simulation

 Table 4.204 mole fractions of streams
Inlet temperature and pressure were selected to minimize the number of plates needed to
achieve desired separation without increasing the utility costs above a reasonable level.
Several simulations with differing inlet temperature and pressure were selected to get the
desired number of plates.

Table 4.205 Stream properties.

From running the simulation a number of times the theoretical number of plates was
determined to be 22.44 with the actual number being 34 with the optimal feed stage of 16.
Condenser and reboiler temperatures were -17°C and 2.9°C respectfully. The heat duty of the
condenser and reboiler is -742kW and 538.3kW respectfully.

Rigorous Column Design.

Rigorous distillation simulation was used in Aspen HYSYS to establish design specifications
using data obtained in the shortcut simulation.
 Figure 4.24 Aspen HYSYS rigorous distillation
Column conditions were input, and product compositions were determined as presented in
table 4.206. The composition of the distillate stream has a DME fraction sufficiently low to
be used for Urea production.

Table 4.206 Stream composition rigorous distillation.

Through rigorous distillation simulation it was established the feed for the stream is 17,
rectifying and stripping sections of the columns are of differing diameter, this was not
considered in the initial design short cut distillation design. Rectifying and stripping sections
have diameter of 0.4595m and 0.5359m respectfully. Sieve trays are used as they are a
versatile cost-effective solution over bubble or valve trays. Sieve trays are usually selected
unless that do not meet the criteria for use, such as with very low vapor flows. Figure 4.25
and 4.26 shows the design specifications for the stripping and rectifying sections.
 Figure 4.24 Design specification for Stripping section

Figure 4.25 Design specification for rectifying section

Based on the above design specifications the downcomer loading varies between 284 m3/h at
plate 34 to 139m3/h at plate 1. Flooding can occur at a flow of 456m3/h so there is sufficient
capacity to make flooding risk unlikely.
 Table 4.207 Design Specification Distillation Column.
Parameter Rectifying Stripping
Column Pressure 303 303
(KPa)
Tray type Sieve Sieve
Tray Spacing (m) 0.6096 0.6096
Plate efficiency 80% 80%
Diameter (m) 0.4595 0.5359
Height (m) 9.754 10.97
Max downcomer 142.8 284.1
loading (m3/h)
Pressure Drop (kPa) 13.4 10.1
Sieve tray hole 12.7 12.7
diameter (mm)
Column internal Stainless steel Stainless steel
material
Wall Thickness (mm) 4.75 4.75

Plate spacing:
Plate spacing is usually between 0.15m to 1m depending on the column diameter. For
columns above 1m in diameter spacing of between 0.3m to 0.6 m, with 0.5m as the initial
estimate and revised as necessary [9]. Rigorous distillation simulation was used to establish
the plate spacing and overall column height. Tray spacing determined to be 0.61m. with
overall column height being 20.73m.

Reboiler Design.
The estimated heat duty of the reboiler was determined using Aspen HYSYS, a duty of
538Kw was estimated. A heating fluid of condensing steam was used to heat the process,
there is a significant amount of steam required for other unit operations within the plant, it
was therefore decided that utilizing steam will minimize the size of the heat exchanger
required saving on capital expenditure. The same design process was undertaken for the heat
exchanger for the reboiler as previously discussed. Sample calculations are presented below.
Heat transfer area was calculated using Eqn 4.101

538 Kw∗1000
A=
W = 5.5m2
850 ∗371.55
m2 K
ΔTlm was calculated using Eqn 4.102. Temperatures of the condensing steam was 373K at 1
atm, the reboiler heated the stream from 273K to 276K in order to vaporize the stream. A
correction factor of 1 was established using a combination of Eqns 4.103-4.105 and Figure
4.11. U was initially estimated to be 850W/m2K using Figure 4.10.
Tube length and diameter were set to 2.44m and 0.101m respectfully with 2 tube passes used.
Equation 4.106 was used to determine the bundle diameter. K1 and n1 were established using
the number of passes and the pitch from data available. [11]
2.2 1/2.291
Db= 0.107( ) = 0.202m
0.156

A fixed head was selected and figure 4.11 was used to establish the clearance between the
bundle and the shell. 0.012m was determined from the graph. Dynamic viscosity, density,
heat capacity, thermal conductivity and Prandtle were found from available literature values
for steam [14]. Equation 4.115 and Figure 4.12 were used to find the Nusselt number and
hence the heat transfer coefficient h. Baffle cut was selected as 35% to try to reduce the
pressure drop across the tube. It is assumed there is negligible difference between the
viscosity at the wall compared to the bulk fluid. Shell equivalent diameter De was found from
equation 4.112.

hoDe
Nu= = (7.5x10-3) (2.56x105) 1.02491/3 (1 ¿0.14 = 7.11x103
Kf
ho= 2.1x105 W/m2K
Equation 4.116 was used to determine the shell side pressure drop. Jf was estimated from
Figure 4.14.

Tube side heat transfer coefficient was determined in a similar was, however it was necessary
to calculate Viscosity and heat capacities based on literature findings. Data available in
Perry’s engineering handbook was used to establish the viscosity of the mixture at the
estimated composition in the reboiler. Equation 4.208 from the handbook was used.
C2 C5
C 1+ +C 3∗ln ( T ) +C 4∗T
T
μ( Pa . s)=e

Where C1-C5 are coefficients and T is temperature in Kelvin.

Viscosity of liquids
C1 C2 C3 C4 C5 μ J/mol.K Fraction
Ammonia -6.743 598.3 -0.7341 -3.67E-27 10 0.00017 0.05
DME -10.62 448.99 0.000083967 0.000127 0.95
CO2 18.775 -402.92 -4.6854 -6.92E-26 10 0.000107 0
Viscosity μ (Pa.s) 1.29E-04

Table 4.208 calculating viscosity of mixture and reboiler composition.

Specific heat capacities were established from mass and energy balance. It was necessary to
convert the units from J/mol.K to J/KgK.
Specific Heat J/mol.K
J/mol.K J/kgK Fraction. Cp
Ammonia 6.18E+03 3.64E+05 0.019 6.91E+03
DME 2.41E+04 5.24E+05 0.981 5.14E+05
CO2 -3.65E+03 -8.30E+04 0 0.00E+00
Cp (J/KgK) 5.21E+05

Table 4.209 Conversion of specific heat capacity

Equations 4.117-4.119 were used to calculate the tube side velocity. Re and Pr number were
calculated . Heat transfer co-efficient was using equation 4.121 and a value of 1.61x104
W/m2K was calculated.
The overall heat transfer coefficient was calculated using eqn 4.123. Data on fouling
estimates were established from literature search [10].
0.023
0.023 x ln ⁡( )
1 1 1 0.021 0.023 1 0.023 1
= + + + x + x
U 2.1 . x 10 3500
5
2 x 15 0.021 2500 0.021 1 x 102

U=851 W/m2K
There is sufficient heat transferred to meet the heat duty of the reboiler at the set design
parameters.
Shell side Tube Side
Fluid Saturated Steam Ammonia, DME mixture
Temperature in (K) 385 273
Temperature out (K) 385 275.9
Pressure Drop (KPa) 4720 117.6
Construction material Steel alloy Stainless steel Sch 40
Area (m2) 1.7
Length (m) 2.44
Number of passes 2
Number of Tubes 2.13
Tube Diameter OD (mm) 0.107
Tube thickness (mm) 6
Shell Diameter (mm) 347
Shell Thickness (mm) 3.2
Baffle Spacing (mm) 138
Pitch Square 1.25Do
Flow type Counter current
Estimate Overall heat transfer Calculated Overall heat
Coefficient transfer Coefficient
W/m2K 850 851
Table 4.210 Reboiler design specifications
Condenser Heat exchanger
Ammonia is used as the refrigerant for the condenser. The Temperature in the condenser is
-17C. To ensure safe operating conditions all piping for the ammonia refrigeration system
must conform to ANSI B31.3, B31.5 and K61.1 standard. Minimum schedule 80 piping is
recommended.
The heat duty of the condenser was established Aspen HYSYS and a duty of -728 kW was
determined. The enthalpy of vaporization of Ammonia is determined using Eqn 4.208.
n
T
(
Hvap= A 1−
Tc ) Eqn 4.208

Where T is the temperature in K, Tc, A and n are regression co-efficients from literature.
0.364
256
Hvap=31.523 1− (
405.65 ) = 21.93 KJ/mol = 1289.8 KJ/Kg

Using figure 4.10 the overall heat transfer coefficient was estimated to be 650W/m2K. The
design procedure for the reboiler was repeated for the condenser. Mass flow of Ammonia
needed to cool the top product to the desired temperature was calculated by Eqn 4.209.
Q= ṁ x hvap
-543 kW = ṁ (Kg/s) * 1289 (KJ/Kg)

ṁ=0.421 Kg/s
Heat transfer Area was found using Eqn. 4.101 to be 29 m2. The calculated overall heat
transfer coefficient was 663 W/m2K indication sufficient heat transfer is achieved.

Shell side Tube Side

Fluid Ammonia refrigerant Ammonia, DME mixture
Temperature in (K) 225 273
Temperature out (K) 225 256
Pressure Drop (KPa) 9.6x103 8132
Construction material Steel alloy Stainless steel Sch 80
Area (m2) 29
Length (m) 3.66
Number of passes 4
Number of Tubes 149
Tube Diameter OD (mm) 0.021
Tube thickness (mm) 2.43
Shell Diameter (mm) 433
Shell Thickness (mm) 3.2
Baffle Spacing (mm) 173
1. Pitch 2. Square 1.25Do
Flow type Counter current
Estimate Overall heat transfer Calculated Overall heat
Coefficient transfer Coefficient
W/m2K 650 660
 Table 4.211 Design Specifications Condenser
6.4

BACKGROUND
An extractive distillation column was simulated with the DWSIM software platform, using
UNIQUAC thermodynamic package to investigate the separation of Dimethyl carbonate-
methanol azeotrope mixture composition with aniline and phenol serving as the entrainer. A
The binary mixture of Dimethyl-carbonate and Methanol forms an azeotrope at 63.73 degrees
Celsius(336.73K) as evidence of the vapour liquid diagram in Fig 1. The implication of an
azeotrope is the equal distribution of species in the vapor and liquid phase. Therefore,
conventional distillation is not feasible in the separation of this mixture. A Known method
for separating an azeotrope system is Extractive distillation. In this method, a third
component (heavy entrainer/solvent) is added to the system. The entrainer/solvent increases
the relative volatility of the two original components, allowing one to go overhead in the
extractive distillation column while the other leaves with the heavy entrainer at the bottom.
To purify the entrainer and recover the Dimethyl carbonate, a second entrainer recovery
column-which is a simple distillation column is used. This entrainer is then recycled back to
the extractive distillation column for further separation. There are a number of other
restrictions to consider when selecting entrainers. One of these restrictions is that the
entrainer must have a boiling point that is significantly different from the other components
in order to allow separation in the second column. Furthermore, when choosing an
entrainer, protection, environmental impact, corrosiveness, cost, and availability should all
be taken into account[ CITATION Mah15 \l 2057 ].

Azeotropic distillation, pressure swing distillation and pervaporation are other known
methods for separating an azeotropic mixture. However, Wang et al investigated pressure-
swing distillation for the methanol-DMC azeotrope and discovered that it is less cost
effective than extractive distillation[ CITATION Kai10 \l 2057 ]. The pervaporation process is
beneficial because it uses less energy, is safer, and is more environmentally friendly than
conventional methods. Nonetheless, since it has yet to be demonstrated in large-scale
applications, because this pervaporation method is constrained in terms of
applications[ CITATION Mah15 \l 2057 ]. Additionally Azeotropic distillation incurs higher
running cost than extractive distillation the entrainer must be vaporized through the top of
the column, thus consuming much energy
Assumption
 Both columns were operated at constant pressure

Aim
 Determining the optimum number of stages
 Determining optimum feed location of binary azeotrope and entrainer
 Determining optimum reflux ratio
 Determining optimum flowrate of the solvent (Entrainer/Feed ratio)
 Determining the suitable candidate Entrainer
Iterative optimization procedure was followed in a bid to achieve the aims listed above
 Methanol-Dimethyl carbonate
P= 101.3 Kpa
365

360

355

350
Temperature

345

340

335

330

325

320
0 0.2 0.4 0.6 0.8 1 1.2
Mole Fraction/Methanol

FIG 4.26: T x-y Diagram for Methanol-DMC

FIG 4.27 LIQUID PHASE COMPOSITION PROFILES

 FIG 4.28 VAPOUR PHASE COMPOSITION PROFILES

Figure 3 shows the variation of vapour phase composition of all the components, along the
column height. The concentrations are shown in mole fractions.
Figure 3 shows an increasing concentration of methanol present in the vapour phase and a
decreasing concentration of DMC. If the azeotrope still persists, then DMC will concurrently
increase alongside methanol in vapour composition. The aniline remains as a liquid
throughout the column as can be seen by its non-existence in the vapour phase in the profile.

FLOWSHEET:
The UNIQUAC Thermodynamic package was chosen as this is recommenced for
highly non-ideal chemical system. This process contains two column. The first is the
extractive distillation column and the second an entrainer recovery column. The azeotropic
mixture with flowrate 28414kg/h and mass composition of 0.814 DMC-0.185 MeoH is fed
into the extractive distillation at the 27th stage while the aniline solvent is fed at the 5th stage.
Its presence on the 5th stage near the top of the column is to ensure significant concentration
in the liquid phase on all trays below. The solvent to feed ratio was set at 0.883. The top
composition and bottom of methanol in the extractive distillation column was set at 0.99 and
0.01while the top and bottom composition of aniline was set at 0.01 and 0.99. Both columns
were operated at constant pressure and pressure drop is neglected. Methanol with 99 percent
purity is recovered in the ED column, DMC with 99 percent purity is produced in the Solvent
recovery column and aniline is recycled back to the ED column with a pure aniline makeup
stream to make up for solvent loss.

Fig 4.28 DWSIM Flowsheet for Extractive distillation

Plates or packing
The decision to make is whether a column is filled packing or with trays. The book of Sinnott and
Towler gives a list of advantages and disadvantages of both types[7]:
 Plate columns can handle a wider range of liquid and gas flow rates
 Packed columns are not suitable for very low liquid flow rates
 The prediction of the efficiency can be more accurate for a plate
 Plate columns can be designed with more assurance than packed columns
 Cooling coils can be placed on a plate
 In plate columns it is easier to withdrawal side streams
 Cleaning a plate column is more easy than packing, only with small diameter columns
packing can be beneficial, than it can be replaced
 For corrosive liquids a packed column is usually cheaper
 Liquid hold-up is lower in a packed column. For safety reasons, this could be important
 Packed column are better to handle foaming mixtures
 Pressure drop per stage can be lower for packed column. For vacuum columns, packing
should always be considered.
 Packing should always be considered for small-diameter columns (<0.6 m), then plates are
difficult to install

The Extractive distillation are plate columns as it operates at atmospheric and have significant liquid
flows(Feed=28414kg/h and solvent =25089kg/h). The assurance of plate column is then preferred
over a packing. Column diameter and height from simulation were computed to be 8.17m and
31.94m
 Fig 4.29

CONDENSER DESIGN

The construction of a shell - tube heat exchangers is quite similar to a shell and tube-heat
exchanger but with significant wider baffle spacing. The condenser is installed in a horizontal
position and its function is to produce a liquid phase of the methanol vapor. The vapor and the
cooling fluid are not in contact and condensation occurs in the shell. The cooling fluid is available
at 25 degree Celsius and 50 degree Celsius.
Vapour flow rate was estimated
D=23030.8
RR = 10.8
L
R=
D
V=L+D
V= 271645.44

Qc from simulation was estimated to be 8.315*E^7, but manual calculation was done to ascertain
this value
Qc is determined by taking a balance over the condenser
 Hv

HL HD
L D

At steady state :
Input = output
Hv = HD + HL + Qc
Hence Qc = Hv- HD - HL
Latent heat of methanol at 64.5 degree Celsius = 1085KJ/Kg
Specific heat capacity = 2.83kj/kg.k
Assuming complete Condesation
Enthalpy of vapour Hv = Latent heat + sensible heat
Hv = 271645.44(0.99*1085)+(64.5-25)
= 2.92*E^8 Kj/h
1kj/h = 0.278W
Hence Hv = 8.1*E^7W

The general equation for heat transfer was used to calculate the overall surface area of the
heat exchanger.

Q=UAΔTm

Where Q is the heat transfer, U is the overall heat transfer co-efficient, A is the surface area
and ΔTm is the mean temperature difference.
 (T 1−t 2)−(T 2−t 1)
ΔTm¿
ln ⁡¿ ¿

T1= Hot fluid inlet temperature

T2= Hot fluid outlet temperature

t1= Cold Fluid inlet temperature

t2= Cold fluid outlet temperature

Rarely are heat exchangers purely co current or counter current, which allows to apply the
LMTD. Most of the them are partially co current and partially counter current which is quite easy
to visualize for a 2 pass tube / 1 pass shell heat exchanger. Therefore, the LMTD must be
corrected by a coefficient F which accounts for those non idealities.

Therefore ΔTm = ΔTlm *Ft

Correction factor was estimated via these equations and the figure 4.11

R= (T 1−T 2)
¿¿
(t 2−t 1 )
S=
(T 1−t 1 )

An assumed Overall Heat coefficient of 950W/m 2 ⃘C was selected from the table below
for Organic vapour with water as cooling fluid

‘
 Figure 4.30

The heat transfer area can then be estimated

Q
A=
T m∗U 0

Tube diameter(do) = 73mm

Tube Inside diameter(di)= 52.26mm
Tube length = 8
Tube pitch(Square Pitch, pt) = 1.25do
Tube side passes = 2
Tube thickness = 7m
Shell thickness = 4.8m alloy steel
To allow sufficient thickness to seal the tubes, the tube sheet thickness should not
be less than the tube outside diameter[8]. Tube thickness is 7m and outside diameter is
0.073m. So this condition is satisfied.
 n1
d
( ) ……..
N t =K 1 b
d0
1 /n 1
N
( ) …..
D b =d 0 t
k1

K1 and n1 constants are retrieved based on number of passes and pitch(Square

pitch)

Nt = Number of tubes
Db= Bundle diameter
Shell Diameter = Db + BDC
The bundle diameter is estimated to be 4.4 m. Therefore a pull-through floating head with100
mm clearance is acceptable.

Db
Nr(Number of tubes in a row)= P
t

TUBE SIDE COEFFICIENT

The cooling fluid water flows through the tube side. An accurate estimate can be made by using equations
developed specifically for water. The physical properties are conveniently incorporated into the correlation. The
following equation has been adapted from data given by Eagle and Ferguson (1930):

4200(1.35+0.02 t) ut 0.8
hi =
di0.2
where
hi -inside coefficient, for water, W/m28C;
t - water temperature, degree C;
ut - water linear velocity, m/s;
di -tube inside diameter, mm.

Q
Cooling water flow =
cp∗∆ T
π
Tubes cross sectional area = ∗di 2
4
Nt
Tubes per pass =
Tube side passes
Total Flow area = Tubes per pass * Tubes cross sectional area

cooling water flow

Water Mass Velocity =
Total flow area
 water mass velocity
Water linear velocity =
density of water
SHELL SIDE COEFFICIENT

( pt−do ) Ds Bs
As=
Pt
The term (pt _ do)/pt is the ratio of the clearance between tubes and the total
distance between tube centers.

pt =tube pitch;
do = tube outside diameter;
Ds = shell inside diameter, m;
lB =baffle spacing, m.

lB = 0.5*Db
Ws
Gs, mass velocity =
As
where
Ws =fluid flow rate on the shell-side, kg/s;

1.27
de = *( pt 2−0.785 do2 ¿
do
de is the shell-side equivalent diameter (hydraulic diameter) For a square pitch arrangement

1
hs de μ
= jh∗ℜ∗Pr 3∗( )
kf μw

μCp
Pr=
k

ρDiu
Re=
μ

For the calculated Reynold’s number, jh was determined from the graph below with baffle
cut of 15%
 Figure 4.31

Overall Transfer Coefficient

do
d o x ln ⁡( )
1 1 1 di d o 1 d o 1
= + + + x + x
U ho h od 2 kw d i hid d i hi
where
Uo = the overall coefficient based on the outside area of the tube, W/m28C;
ho = outside fluid film coefficient, W/m28C;
hi =inside fluid film coefficient, W/m28C;
hod = outside dirt coefficient (fouling factor), W/m28C;
hid = inside dirt coefficient, W/m28C;
kw = thermal conductivity of the tube wall material, W/m8C;
di = tube inside diameter, m;
do = tube outside diameter, m.

Shell Side Tube side

Fluid Methanol Vapour Water
Temperature in ( ⃘C) 64.5 25
Temperature out ( ⃘C) 64.5 50
Construction Materials Alloy steel Cupro-nickel alloy
Area(m2)
3418
Length(m)
 8
Number of passes
2
Tube Diameter (Do)mm
73
Tube Inside Diameter Di(mm)
52
Shell thickness(m)
4.8
Baffle spacing
2237
Pitch
Square1.25D0
Flow Type

Estimated Overall Transfer Calculated Overall Transfer

coefficient coefficient
W/m2.K 950 934
Table 4.212 Design specifications for condenser

References:

1. Sinnott, R.K., Chemical engineering design. 4th ed. ed. Coulson & Richardson's chemical
engineering ; 6, ed. J.M. Coulson and J.F. Richardson. 2005, Oxford: Elsevier Butterworth-
Heinemann.
2. Towler, G. and R. Sinnott, Chemical Engineering Design: Principles, Practice and Economics of
Plant and Process Design. 2012, Oxford: Elsevier Science & Technology.
3. Shawabkeh, R., Handout: Step-by-step for Heat Exchanger design. 2015.
4. Don, W.G. and H.P. Robert, Perry's Chemical Engineers' Handbook, Eighth Edition. 8th ed. /
ed. 2008, New York: McGraw-Hill Education.
5. Yaws, C.L., Chemical properties handbook : physical, thermodynamic, environmental,
transport, safety, and health related properties for organic and inorganic chemicals. 1999,
New York ;: McGraw-Hill.
6. Wagner, W. and H.-J. Kretzschmar, International Steam Tables - Properties of Water and
Steam Based on the Industrial Formulation IAPWS-IF97: Tables, Algorithms, Diagrams, and
CD-ROM Electronic Steam Tables - All of the Equations of IAPWS-IF97 Including a Complete
Set of Supplementary Backward Equations for Fast Calculations of Heat Cycles, Boilers, and
Steam Turbines. 2007, Berlin, Heidelberg: Springer Berlin / Heidelberg.
7. Sinnott, R. and G. Towler, Chemical Engineering Design. 5th ed. 2009, Oxford: Elsevier. 1255.
8. 1. State-of-the-Art Technologies for Separation of Azeotropic Mixtures. Mahdi Taha, Ahmad
Arshad, Nasef Mohamed, Ripin Adnan. UK : Taylor & Francis, 2015, Vol. 44.

2. Design and Control of Dimethyl Carbonate-Methanol Separation via Extractive in Dimethlyl

Carbonate Reactive- distillation process. Kai-Yi Hsu, Yuan-Chang Hsiao, and I-Lung Chien. Taiwan :
s.n., 2010, Vol. 49.

3. Gavin Towler, Ray Sinnot. Chemical Engineering Design Principles, Practice and. United states :
Butterworth-Heinemann, 2008.
1. Sinnott, R.K., Chemical engineering design. 4th ed. ed. Coulson & Richardson's chemical
engineering ; 6, ed. J.M. Coulson and J.F. Richardson. 2005, Oxford: Elsevier Butterworth-
Heinemann.
2. Towler, G. and R. Sinnott, Chemical Engineering Design: Principles, Practice and Economics of
Plant and Process Design. 2012, Oxford: Elsevier Science & Technology.
3. Shawabkeh, R., Handout: Step-by-step for Heat Exchanger design. 2015.

References

1. State-of-the-Art Technologies for Separation of Azeotropic Mixtures. Mahdi Taha, Ahmad Arshad,
Nasef Mohamed, Ripin Adnan. UK : Taylor & Francis, 2015, Vol. 44.

2. Design and Control of Dimethyl Carbonate-Methanol Separation via Extractive in Dimethlyl

Carbonate Reactive- distillation process. Kai-Yi Hsu, Yuan-Chang Hsiao, and I-Lung Chien. Taiwan :
s.n., 2010, Vol. 49.

3. Gavin Towler, Ray Sinnot. Chemical Engineering Design Principles, Practice and. United states :
Butterworth-Heinemann, 2008.