Advanced Separation Technology

(6CPT40)

Rate based models for separations

by phase addition/creation

Prof.dr.ir. Martin van Sint Annaland

STW 0.39, [email protected], 040-2472241
 Separation by phase addition/creation
• Discussed here for distillation
completely analogous for absorption, stripping, extraction, etc.

Chemical Engineering and Chemistry 30-1-2024 Page 1

 Separation by phase addition/creation

• Distillation is a mature technology

Use Asymptote Distillation
Absorption
Crystallization
Extraction
Ion Exchange Extractive&
Azeotropic
Adsorption: Distillation
What Process gas feed
Engineers Use
Membranes
Adsorption:
liquid feed

First Application
Invention Technology
Asymptote
What Process Engineers Know
Chemical Engineering and Chemistry 30-1-2024 Page 2
 Separation by phase addition/creation

• Separation column design strategy

– Set product specifications (degree of separation required)
– Set operating conditions (viz. operating pressure)
– Select type of contacting device: plates vs. packing
– Determine the stages and reflux requirements
(number of equilibrium stages)
– Design column internals
– Mechanical design (vessel & internal fittings)

• Stage and reflux requirements

– Short-cut methods: McCabe-Thiele/Fenske-Underwood-Gilliland
– Rigorous methods:
– Equilibrium-based models
– Rate-based models
Chemical Engineering and Chemistry
 Overview

• Basics of distillation of an ideal binary mixture

– Flash calculation
– McCabe-Thiele method
– Fenske-Underwood-Gilliland (FUG) short-cut method

• Extension FUG method

– Extension to multi-component systems
– Extension with stage efficiencies

• Extension to non-ideal systems

• Rigorous models
– Equilibrium models
– Rate based models
Chemical Engineering and Chemistry 30-1-2024 Page 4
 Basics of ideal binary distillation

• Continuous distillation of a binary mixture A+B

• A more volatile than B
Cooling water
outlet
Heat removal

Feed:
Mixture of
A and B
Cooling water
inlet Distillate:
Boiling liquid
mixture More A and less B
than in the feed

Heat supply Bottom product: Distillation: separation

Less A and more B on the basis of
differences in volatility
than in the feed

Chemical Engineering and Chemistry 30-1-2024 Page 5

 Basics of ideal binary distillation

• Vapour-liquid equilibrium description for ideal fluids

Dalton: 𝑃𝑃𝐴𝐴 = 𝑦𝑦𝐴𝐴 𝑃𝑃 Combination of Dalton & Raoult’s laws:

Raoult: 𝑃𝑃𝐴𝐴 = 𝑥𝑥𝐴𝐴 𝑃𝑃𝐴𝐴𝑠𝑠 𝑃𝑃𝐴𝐴 𝑠𝑠
𝑦𝑦𝐴𝐴 𝑃𝑃
= 𝐾𝐾𝐴𝐴 = 𝑃𝑃 =
𝐴𝐴
⇒ 𝑃𝑃𝐴𝐴𝑠𝑠 = saturation
𝑥𝑥𝐴𝐴 𝑃𝑃𝐴𝐴 𝑃𝑃 pressure
𝑠𝑠
Dalton: 𝑃𝑃𝐴𝐴 = 𝑦𝑦𝐴𝐴 𝑃𝑃 𝑃𝑃𝐴𝐴
Ideal gas law:
𝑃𝑃𝐴𝐴 𝑉𝑉 = 𝑛𝑛𝐴𝐴 𝑅𝑅𝑅𝑅 𝑃𝑃𝐴𝐴 𝑛𝑛𝐴𝐴 Separation factor αAB
⇒ = = 𝑦𝑦𝐴𝐴
𝑃𝑃𝑃𝑃 = 𝑛𝑛𝑛𝑛𝑛𝑛 𝑃𝑃 𝑛𝑛 𝑦𝑦𝐴𝐴 𝑃𝑃𝐴𝐴𝑠𝑠
𝑥𝑥𝐴𝐴 𝑃𝑃 𝑃𝑃𝐴𝐴𝑠𝑠
𝛼𝛼𝐴𝐴𝐴𝐴 = = 𝑠𝑠 = 𝑠𝑠 = 𝑓𝑓(𝑇𝑇)
Raoult: 𝑃𝑃𝐴𝐴 = 𝑥𝑥𝐴𝐴 𝑃𝑃𝐴𝐴𝑠𝑠 𝑦𝑦𝐵𝐵 𝑃𝑃𝐵𝐵 𝑃𝑃𝐵𝐵
Interactions between A and B 𝑥𝑥𝐵𝐵 𝑃𝑃
indentical, then for a binary mixture: α = relative volatility
𝑃𝑃 = 𝑥𝑥𝐴𝐴 𝑃𝑃𝐴𝐴𝑠𝑠 + 𝑥𝑥𝐵𝐵 𝑃𝑃𝐵𝐵𝑠𝑠
Chemical Engineering and Chemistry 30-1-2024 Page 6
 Basics of ideal binary distillation

• Equilibrium of binary mixture: (Notation: X for xA and Y for yA)

Equilibrium curve = relation between the composition of a liquid
𝛼𝛼𝛼𝛼 and the vapour in equilibrium with it
𝑌𝑌 =
𝛼𝛼 − 1 𝑋𝑋 + 1

Derivation equilibrium curve:

50
𝑌𝑌
𝑋𝑋 𝑌𝑌 1 − 𝑌𝑌 20
𝛼𝛼 = ⇒ = 𝛼𝛼
1 − 𝑌𝑌 𝑋𝑋 1 − 𝑋𝑋 10
1 − 𝑋𝑋 Y=X
2
⇒ 𝑌𝑌 1 − 𝑋𝑋 = 𝛼𝛼𝛼𝛼 1 − 𝑌𝑌
α =1
⇒ 𝑌𝑌 1 − 𝑋𝑋 + 𝛼𝛼𝛼𝛼 = 𝛼𝛼𝛼𝛼
𝛼𝛼𝛼𝛼 A is volatile comp. ⇒ α >1
⇒ 𝑌𝑌 =
𝛼𝛼 − 1 𝑋𝑋 + 1
Chemical Engineering and Chemistry 30-1-2024 Page 7
 Basics of ideal binary distillation

• Flash calculation: D
YD
F, B, D: molar flow rates [kmol/s]
XF, XB, YD: mole fractions [-]
YD Cooling water
Distillate
F
Feed XB Given: F, XF, α
XF
steam
steam Required: B, D, XB, YD
B
Bottom product
XB
Total material balance: 𝐹𝐹 = 𝐵𝐵 + 𝐷𝐷
Ratio
4 B/D is
unknowns
Comp. material balance: 𝑋𝑋𝐹𝐹 𝐹𝐹 = 𝑋𝑋𝐵𝐵 𝐵𝐵 + 𝑌𝑌𝐷𝐷 𝐷𝐷 determined by
3 equations?
𝛼𝛼𝑋𝑋𝐵𝐵 energy supply
Equilibrium: 𝑌𝑌𝐷𝐷 =
𝛼𝛼 − 1 𝑋𝑋𝐵𝐵 + 1
Chemical Engineering and Chemistry 30-1-2024 Page 8
 Basics of ideal binary distillation

• Example flash calculation: Benzene/toluene separation

Given: F = 1 kmol/s; XF = 0.50; α = 2.5; B/D = 3
F= B + D
 B B α XB
X=
FF X B B + YD D ⇒ 1+ D  X F − X B D =
YD =
  (α − 1) X B + 1
α XB
YD =
(α − 1) X B + 1
Analytical solution: Enriched in
XB 5 benzene
2 − 3 XB = 2
YD= = 0.667
2

2 XB + 1
3 3

= D 0.25 kmol ⋅ s−1

(2 − 3 XB ) ( 3
2
X B + 1) − X B =
5
2
0
⇒
3 X B + 2 − 92 X B2 − 3 X B − 52 X B =0 X B= 94= 0.444
Reduced
2
X + XB − =
5
0 4 =B 0.75 kmol ⋅ s−1 in benzene
B 9 9
Chemical Engineering and Chemistry 30-1-2024 Page 9
 Basics of ideal binary distillation

• Example flash calculation: Benzene/toluene separation

Given: F = 1 kmol/s; XF = 0.50; α = 2.5; B/D = 3

 B B α XB slope = –B/D
YD = 1 +
 D F X − X =
D (α − 1) X B + 1
B
  1
feed-line
feed-line 0.8
Equilibrium 0.67
curve
0.6 Equilibrium
curve
Grafical solution: Y
0.4
Intersection of equilbrium 0.2
curve and feed-line
0 0.44
0 0.2 0.4 0.6 0.8 1
X
Chemical Engineering and Chemistry 30-1-2024 Page 10
 Basics of ideal binary distillation

• Example flash calculation: Benzene/toluene separation

Effect of B/D (= energy supply): Subcooled
liquid
B/D=2 B/D=∞
1
B/D=1 Liquid at
half vaporized boiling
point
0.8 B/D=½

0.6
B/D=0 Energy supply
determines slope
Y

Vapour at
0.4 dew point q-line!

Superheated
vapour 0.2

(XF = 0.50, α = 2.5)

0
0 0.2 0.4 0.6 0.8 1
Chemical Engineering and Chemistry 30-1-2024 Page 11
X
 Basics of ideal binary distillation

• From flash drums to distillation column

Chemical Engineering and Chemistry 30-1-2024 Page 12

 Basics of ideal binary distillation

Distillation
column creates
itself the two
Distillate
phases required
Reflux
for the counter
process

Feed
Tray
columns
vs.
packed
columns

Reboiler =
tray nr. 1 Bottom product
Chemical Engineering and Chemistry 30-1-2024 Page 13
 Basics of ideal binary distillation
• Tray columns: types of trays

sieve
tray

bubble
cap tray

valve
tray
Chemical Engineering and Chemistry 30-1-2024 Page 14
Basics of ideal binary distillation
• Tray columns: types of trays

Create intimate contact

between gas and liquid in
‘cross flow’ on the tray

30-1-2024
Basics of ideal binary distillation
• Tray columns: comparison of different tray types
• Column dimensioning:
− Height of the column:
related to number of
trays required for a
target purity:
− Number of trays
− Tray efficiency
− Distance of trays

− Diameter of the column

required for a given
capacity:
− Tray operating region
− Vapor flow rate
− Liquid flow rate

30-1-2024
 Basics of ideal binary distillation

• Packed columns: types of packing

Random (bulk) packing Structured packing

Chemical Engineering and Chemistry 30-1-2024 Page 17

 Basics of ideal binary distillation
• Packed columns: other internals
− Liquid (re)distributers (every 6 m)
− Gas distributers
− Liquid collectors
− Wall wipers
− Support grid + hold down plates

Chemical Engineering and Chemistry 30-1-2024 Page 18

 Basics of ideal binary distillation

• Packed columns: comparison of different packings

Random (bulk) packing Structured packing

− Cheap − High costs

− Inhomogeneous − Homogeneous and
− High pressure drop good V/L distribution
− Low pressure drop
Chemical Engineering and Chemistry 30-1-2024 Page 19
Basics of ideal binary distillation

• Packed columns: comparison of different packings

30-1-2024
 Basics of ideal binary distillation

• Tray columns vs. packed columns

Chemical Engineering and Chemistry 30-1-2024

30-1-2024 Page 21
 Basics of ideal binary distillation

• Different reboilers

Chemical Engineering and Chemistry 30-1-2024 Page 22

 Basics of ideal binary distillation

• Column dimensioning

• Height of the column:

Related to the number of stages required for
a target purity:
− Number of stages
− Tray efficiency
− Distance between the trays

• Diameter of the column:

− Tray operating region
− Vapor flow rate
− Liquid flow rate

Chemical Engineering and Chemistry 30-1-2024 Page 23

 Basics of ideal binary distillation

• Height of a tray column

𝐻𝐻 = 𝐻𝐻𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 𝑁𝑁𝑠𝑠

Hspacing = space between two trays ⇒ From empirical correlation

(depends on type of trays)
Ns = total number of trays
Hspacing = 0.2-0.8 ≈ 0.5 m
= Neq/Eov
⇒ Calculated from theoretical number of stages
and overall column efficiency

⇒ Modelling of gas / liquid contact is required

Neq: short-cut method via McCabe-Thiele/FUG method

Eov: tray models

Chemical Engineering and Chemistry 30-1-2024

30-1-2024 Page 24
 Basics of ideal binary distillation

• McCabe-Thiele method:
YD
Total material balance: F= D + B
D
X D = YD MVC material balance: FX
= F DX D + BX B

For the design of a distillation column:

XF
Given: F en XF
F Specified: XD en XB

XF − XB
D= F
XD − XB Distillate & bottom
⇒ products fixed by
XD − XF
B= F material balances
B XD − XB
XB
Chemical Engineering and Chemistry 30-1-2024 Page 25
 Basics of ideal binary distillation

• McCabe-Thiele method: Constant Molar Overflow

V” V” For ideal mixtures:
L”
vapour and liquid molar
V” L”
L” flow rates are constant,
V” L”
but composition is changing
V” L”
V” L” Rectifying section:
V” and L” are constant
V” L” RECTIFI-
CERENDE
SECTIE
V” L”
F F Feed-stage:
V’ L’ V” and L” change to V’ and L’
STRIP-
PENDE
V’ L’ SECTIE
V’ L’
V’ L’
Stripping section:
V’ and L’ are constant
V’ L’
V’ V’

L’
Chemical Engineering and Chemistry L’
30-1-2024 Page 26 V’, V”, L’, L”: molar flow rates [mol/s]
 Basics of ideal binary distillation

• McCabe-Thiele method: stripping section

V’ L’
Yi Xi+1 Total material balance: L=′ V ′ + B
MVC material balance: L′X=
i +1 V ′Yi + BX B
i

⇒ L′X i +1 = V ′Yi + ( L′ − V ′ ) X B

⇒ V ′Y=
i L′X i +1 + (V ′ − L′ ) X B
Y2
2 V’, YB
L′  L′  Operating curve
⇒ =
Yi X i +1 +  1 −  X B
Reboiler V′  V′ stripping section
L
X2
B L′
XB N.B. Equal molar overflow: = constant
V′
L′
MVC = Most volatile component >1
Chemical Engineering and Chemistry 30-1-2024 Page 27 V′
 Basics of ideal binary distillation

• McCabe-Thiele method: Operating Y= L′ X +  1 − L′  X

curve: V′  V′ B
stripping section  
Y L′
slope: >1
V′
V’ Equilibrium
L’
Y4 Y4 curve
Y3
X4 4
Y2
Y3

X3 3 YB

Y2

X2 2
V’ XB
YB
YB
1 XB X2 X3 X 4 X
L’ XB  L′ 
1 −
 V ′  XB
X2
Chemical Engineering and Chemistry Reboiler B
30-1-2024 Page 28
 
 Basics of ideal binary distillation

• McCabe-Thiele method: rectifying section

V” Total material balance: ′′ L′′ + D

V=
YN
MVC material balance: V ′′=
Yi −1 L′′X i + DX D
N
YN-1 XN

L” D ⇒ V ′′Yi −1 = L′′X i + (V ′′ − L′′ ) X D

X0=XD XD

L′′  L′′ 
Yi ⇒ Yi=
−1 X i + 1−  X D
i V ′′  V ′′ 

V” L” L′′
Yi-1 Xi N.B. Equal molar overflow: = constant
V ′′
L′′
MVC = Most volatile component <1
V ′′
Chemical Engineering and Chemistry 30-1-2024 Page 29
 Basics of ideal binary distillation

• McCabe-Thiele method: Operating Y

=
L′′  L′′ 
X + 1−  X D
rectifying section curve: V ′′  V ′′ 
V’’ YN
Condenser

YN −1
V’’ L’’ D Equilibrium
YN curve
L’’ XD XD
YN −2
XN N

YN-1 YN −3
Y
XN-1 N-1

YN-2

L′′  L′′ L′′ R

XN-2 N-2  slope: <1
1 −
 V ′′  X D V ′′ =
  V ′′ R + 1
X X N −2 X N −1 XN XD
Chemical Engineering and Chemistry 30-1-2024
Page 30
 Basics of ideal binary distillation

• Definition reflux ratio R

V” External L′′
Reflux ratio: R=
D
D
L” Internal L′′ R
Reflux ratio: =
V ′′ R + 1
L”
= R = uitwendige
External reflux ratio:
V” L” D refluxverhouding
L′′
L′′ L′′ D R
= = =
L” V ′′ L′′ + D L′′ + 1 R + 1
= inwendige refluxverhouding
Internal reflux ratio
V” D

Chemical Engineering and Chemistry 30-1-2024 Page 31

 Basics of ideal binary distillation

• Feed stage considerations:

V” L” How the feed is introduced is a

design consideration!

Partially vaporized (FV)

or partially condensed (FL)
F = FL + FV FV
V” = V’ + FV
FL

L’ = L” + FL

Usually: feed as liquid at

boiling point (FL = F, FV = 0)
V’ L’

Chemical Engineering and Chemistry 30-1-2024 Page 32

 Basics of ideal binary distillation

• feed-line
The feed-line (or q-line) intersects the operating lines of
the stripping and rectifying sections:

V”,yf L” Rect. section: ′′Y L′′X + DX D

V=
′Y L′X − BX B
Stripping section: V=
Feed
f
F, xF Feed stage
Subtract: (V ′′ − V ′ )Y =( L′′ − L′ ) X + DX D + BX B
= FX F
V’ L’,xf
⇒ −FL X + ( FV + FL ) X F
FVY =
V” = V’ + FV
feed-line:
⇒ F  F 
L’ = L” + FL Y= − L X + 1+ L  XF
FV  FV 

Chemical Engineering and Chemistry 30-1-2024 Page 33

 Basics of ideal binary distillation

• Overview operating curves:

Rectifying section:
Equilibrium
curve rectif. L′′  L′′ 
feed-line Y= X + 1−  X D
V ′′  V ′′ 
Y
Stripping section:
L′  L′ 
Y= X + 1−  X B
V′  V′
feed-line:
strip. FL  FL 
Y =− X + 1+  XF
 L′′  FV  FV 
1 −
 V ′′  X D
 
XB XF X XD L′′ R
=
V ′′ R + 1
 L′ 
1 −
Chemical  X B and Chemistry
 V  ′
Engineering 30-1-2024 Page 34
 Basics of ideal binary distillation
1
Equilibrium curve
2

feed-line

McCabe Thiele:
Y (vapour mole fraction)

3
determine number of
equilibrium stages
Operating curve
rectifying section

4
number of stages
= number of steps – 1
(because of reboiler)
Reboiler

Operating curve
stripping section

XB XF XD
X (liquid mole fraction)

Chemical Engineering and Chemistry 30-1-2024 Page 35

 Basics of ideal binary distillation

• Summary design using McCabe-Thiele method:

− Determine product flows B and D using specifications (XD and XB)
− Draw equilibrium curve in XY-diagram
− Calculate/chose slope of feed-line (FL/FV) and draw feed-line with
this slope from (XF,XF)
− Chose reflux ratio R
− Calculate y-axis intersection of operating curve of the rectifying
section: xD/(R+1) and draw curve through this and point (XD,XD)
− Draw operating curve of the stripping section by connecting
intersection feed-line and operating curve rectifying section with
point (XB,XB)
− Draw steps and determine number of equilibrium stages

How to select reflux ratio R ?

Chemical Engineering and Chemistry 30-1-2024 Page 36
 Basics of ideal binary distillation

• Number of stages vs. reflux

1.0 1.0 6
9 5
0.9 8 0.9 4
7
0.8 6 0.8
5 3
0.7 0.7
4
0.6 0.6
3 2
0.5 0.5
Y

Y
0.4
2 0.4
0.3 1 0.3
1
0.2 R 0.2
R
0.1 R = 1.2 0.1 R = 3.0
0.0 0.0
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
X X

R = 1.2 ⇒ Reboiler + 9 stages R = 3.0 ⇒ Reboiler + 6 stages

Chemical Engineering and Chemistry 30-1-2024 Page 37

 Basics of ideal binary distillation

• Number of stages vs. reflux

Minimum
Reflux Optimal reflux?
Actual number of
stages required?

Short-cut methode:
Minimum
number
Fenske-Underwood-
Nr of of stages Gilliland (FUG)
stages

Reflux

Chemical Engineering and Chemistry 30-1-2024 Page 38

 Basics of ideal binary distillation

• Fenske-Underwood-Gilliland short-cut method:

Minimal reflux-ratio Rmin (Underwood)
Rmin is determined
by the “Pinch”
P
 L′′  X D − YP
YP
 V ′′  = X − X
 min
Y (vapour mole fraction)

D P

“Pinch”  L′′ 
R = Rmin  V ′′ 
N=∞ Rmin =  min
 L′′ 
1−  
 V ′′ min

N.B. Minimal reflux

XB XP XF XD ⇒ infinite number of stages
X (liquid mole fraction)

Chemical Engineering and Chemistry 30-1-2024 Page 39

 Basics of ideal binary distillation

• Fenske-Underwood-Gilliland short-cut method:

Optimal reflux ratio Ropt:

Total Costs
Distillation Costs

Rule of thumb: Ropt ≈ 1.2 Rmin

Energy

Factor (1.1-1.5) depends on:

– energy costs,
– condenser cooling fluid
– construction materials, etc.
Economic
optimum
Investment

0
0 RMIN ROPT Reflux ratio, R
Chemical Engineering and Chemistry 30-1-2024 Page 40
 Basics of ideal binary distillation

• Fenske-Underwood-Gilliland short-cut method:

Minimum number of stages Nmin (Fenske):
Minimum number of stages
65
54
1.0
at totale reflux (R→∞):
0.9 34
⇒ Stripping and rectifying 0.8
3
0.7
operating curves identical 2
0.6
to diagonal 2
0.5

Y
0.4
Fenske equation:
0.3
11
𝑋𝑋𝐷𝐷 1 − 𝑋𝑋𝐵𝐵
ln 0.2
R
𝑋𝑋𝐵𝐵 1 − 𝑋𝑋𝐷𝐷 0.1
𝑁𝑁𝑚𝑚𝑚𝑚𝑚𝑚 =
ln 𝛼𝛼𝑎𝑎𝑎𝑎𝑎𝑎 0.0
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
X
Min. number of stages
including reboiler
R = ∞ ⇒ Reboiler + 5 stages
Chemical Engineering
Chemical andand
Engineering Chemistry
Chemistry 30-1-2024
30-1-2024 Page 41 41
Page
 Basics of ideal binary distillation

• Fenske-Underwood-Gilliland short-cut method:

Gilliland correlation

𝑁𝑁 − 𝑁𝑁𝑚𝑚𝑚𝑚𝑚𝑚
=
𝑁𝑁 + 1
1 + 54.4𝐺𝐺 𝐺𝐺 − 1
1 − exp
11 + 117.2𝐺𝐺 𝐺𝐺 0.5

𝑅𝑅 − 𝑅𝑅𝑚𝑚𝑚𝑚𝑚𝑚
𝐺𝐺 =
𝑅𝑅 + 1

Note: N includes (partial) reboiler and (partial) condenser, if used

Chemical Engineering and Chemistry 30-1-2024 Page 42

 Basics of ideal binary distillation

• Fenske-Underwood-Gilliland short-cut method:

− Determine Rmin via Pinch point (Underwood)

− Chose Ropt (≈ 1.2 Rmin)
− Determine Nmin using Fenske’s equation
− Use Gilliland’s correlation to determine N

Chemical Engineering and Chemistry 30-1-2024 Page 43

 Basics of ideal binary distillation

• Limitations of McCabe-Thiele method:

• Binary mixture
• Ideal mixture (constant relative volatility)
• Constant molar overflow
• All stages at equilibrium

• Modifications for multicomponent mixtures:

• Light key/Heavy key
• Tray efficiencies
• Non-ideal mixtures
⇒ Use computer-simulations!

Chemical Engineering and Chemistry 30-1-2024 Page 44

 Extended Fenske-Underwood-Gilliland method

• Extension for multicomponent systems

Assumptions:
− System is ideal
− Constant molar overflow (as in McCabe Thiele method)
− Separation is essentially taking place between a
light key (LK) component and a heavy key (HK) component
LK = heaviest component in the top
HK = lightest component in the bottom

Chemical Engineering and Chemistry 30-1-2024 Page 45

 Extended Fenske-Underwood-Gilliland method

• Extension for multicomponent systems

Fenske’s equation for Nmin:
𝑥𝑥𝐿𝐿𝐿𝐿 𝑥𝑥𝐻𝐻𝐻𝐻
log
𝑥𝑥𝐻𝐻𝐻𝐻 𝐷𝐷
𝑥𝑥𝐿𝐿𝐿𝐿 𝐵𝐵
𝑁𝑁𝑚𝑚𝑚𝑚𝑚𝑚 =
log 𝛼𝛼𝐿𝐿𝐿𝐿

𝐾𝐾𝐿𝐿𝐿𝐿 𝐾𝐾𝐿𝐿𝐿𝐿 𝐾𝐾𝑖𝑖 𝑃𝑃𝑖𝑖𝑠𝑠

𝛼𝛼𝐿𝐿𝐿𝐿 = 𝛼𝛼𝑖𝑖 = 𝐾𝐾𝑖𝑖 =
𝐾𝐾𝐻𝐻𝐻𝐻 𝐾𝐾𝐻𝐻𝐻𝐻 𝐾𝐾𝐻𝐻𝐻𝐻 𝑃𝑃
𝐷𝐷 𝐵𝐵

Average relative Relative volatility of for ideal gas

volatility of the light component i with respect and liquid
key with respect to to the heavy key mixtures
the heavy key

Chemical Engineering and Chemistry 30-1-2024 Page 46

 Extended Fenske-Underwood-Gilliland method

• Extension for multicomponent systems

Underwood’s equation for Rmin:
𝐶𝐶
𝛼𝛼𝑖𝑖 𝑥𝑥𝑖𝑖,𝐷𝐷
𝑅𝑅𝑚𝑚𝑚𝑚𝑚𝑚 =� −1
𝛼𝛼𝑖𝑖 − 𝜃𝜃
𝑖𝑖=1
𝐶𝐶
𝛼𝛼𝑖𝑖 𝑥𝑥𝑖𝑖,𝐹𝐹
where θ is the solution of: � = 1 − 𝑞𝑞
𝛼𝛼𝑖𝑖 − 𝜃𝜃
𝑖𝑖=1

heat to vaporize 1 mole of feed

and q is defined as: 𝑞𝑞 =
molar latent heat of feed
(For liquid feed at boiling point: q = 1)

Chemical Engineering and Chemistry 30-1-2024 Page 47

 Extended Fenske-Underwood-Gilliland method

• Extension for multicomponent systems

Feed point location by empirical correlation Kirkbride:
2
𝑁𝑁𝑅𝑅 𝐵𝐵 𝑥𝑥𝐹𝐹,𝐻𝐻𝐻𝐻 𝑥𝑥𝐵𝐵,𝐿𝐿𝐿𝐿
log = 0.206 log (not very accurate!)
𝑁𝑁𝑆𝑆 𝐷𝐷 𝑥𝑥𝐹𝐹,𝐿𝐿𝐿𝐿 𝑥𝑥𝐷𝐷,𝐻𝐻𝐻𝐻

NR = number of stages above the feed, including condenser

NS = number of stages below the feed, including reboiler

Chemical Engineering and Chemistry 30-1-2024 Page 48

 Extended Fenske-Underwood-Gilliland method

• Extension for stage efficiencies

Real effect
• Efficiency =
Maximum (equilibrium) effect
𝑦𝑦𝑖𝑖,𝑛𝑛 − 𝑦𝑦𝑖𝑖,𝑛𝑛+1
• Murphree vapor efficiency for stage n: 𝐸𝐸𝑀𝑀𝑀𝑀 = ∗
𝑦𝑦𝑖𝑖,𝑛𝑛 − 𝑦𝑦𝑖𝑖,𝑛𝑛+1
operating
curve
equilibrium
curve

Chemical Engineering and Chemistry 30-1-2024

30-1-2024 Page 49
Extended Fenske-Underwood-Gilliland method

• Extension for stage efficiencies

Approach to equilibrium
• Stage efficiency:
𝑦𝑦𝑖𝑖,𝑛𝑛 − 𝑦𝑦𝑖𝑖,𝑛𝑛+1
𝐸𝐸𝑀𝑀𝑀𝑀 = ∗
𝑦𝑦𝑖𝑖,𝑛𝑛 − 𝑦𝑦𝑖𝑖,𝑛𝑛+1

• Overall efficiency
of the column:
𝑁𝑁𝑒𝑒𝑒𝑒
𝐸𝐸𝑜𝑜𝑜𝑜 =
𝑁𝑁

• What is de relation
between EMV and EO?

30-1-2024
 Extended Fenske-Underwood-Gilliland method

• Extension for stage efficiencies

Overall efficiency of a tray column:

Distillation:
Eov = 50-80%
Absorption: Empirical correlation by
Eov = 10-30% Lockhart and Leggett

Chemical Engineering and Chemistry 30-1-2024

30-1-2024 Page 51
 Extended Fenske-Underwood-Gilliland method

• Extension for stage efficiencies

Overall (stage) efficiency correlations:

Drickamer and Bradford: 𝐸𝐸𝑜𝑜𝑜𝑜 = 19.2 − 57.8 log 𝜇𝜇𝐿𝐿

(0.2 < 𝜇𝜇𝐿𝐿 < 1.6 𝑐𝑐𝑐𝑐)
−0.226
Lockhart and Leggett: 𝐸𝐸𝑜𝑜𝑜𝑜 = 50.3 𝛼𝛼𝐿𝐿𝐿𝐿,𝐻𝐻𝐻𝐻 𝜇𝜇𝐿𝐿

O’Connell: 𝐾𝐾𝑀𝑀𝐿𝐿 𝜇𝜇𝐿𝐿

log 𝐸𝐸𝑜𝑜𝑜𝑜 = 1.597 − 0.199 log
𝜌𝜌𝐿𝐿
2
𝐾𝐾𝑀𝑀𝐿𝐿 𝜇𝜇𝐿𝐿
−0.0896 log
𝜌𝜌𝐿𝐿

In general: not very accurate, nor generally applicable!

Chemical Engineering and Chemistry 30-1-2024

30-1-2024 Page 52
 Non-ideal systems

• Extension for non-ideal mixtures

For each phase in a multi-phase multicomponent system:
Ni = number of
Total Gibbs free energy: 𝐺𝐺 = 𝐺𝐺 𝑇𝑇, 𝑃𝑃, 𝑁𝑁1 , 𝑁𝑁2 , … , 𝑁𝑁𝐶𝐶 moles of species i

At equilibrium: the total G for all phases is at its minimum

⇒ “free-energy minimization techniques”
𝐶𝐶
Total differential: 𝑑𝑑𝑑𝑑 = −𝑆𝑆𝑆𝑆𝑆𝑆 + 𝑉𝑉𝑉𝑉𝑉𝑉 + ∑𝑖𝑖=1 𝜇𝜇𝑖𝑖 𝑑𝑑𝑁𝑁𝑖𝑖
For a closed system with 𝐶𝐶 𝑝𝑝 𝑝𝑝
𝑑𝑑𝐺𝐺𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 = ∑𝑁𝑁 ∑ 𝑖𝑖=1 𝑖𝑖 𝑑𝑑𝑁𝑁𝑖𝑖
𝜇𝜇
two or more phases: 𝑝𝑝=1
𝑃𝑃,𝑇𝑇
Conservation of moles 𝐶𝐶 𝑝𝑝 (1) 𝑝𝑝
𝑑𝑑𝐺𝐺𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 = ∑𝑁𝑁 ∑ 𝜇𝜇 − 𝜇𝜇 𝑑𝑑𝑁𝑁𝑖𝑖
(absence of chem. rxns) 𝑝𝑝=2 𝑖𝑖=1 𝑖𝑖 𝑖𝑖
𝑃𝑃,
(1) 𝑝𝑝
𝑑𝑑𝑁𝑁𝑖𝑖 = − ∑𝑁𝑁
𝑝𝑝=2 𝑑𝑑𝑁𝑁𝑖𝑖 1 2 𝑁𝑁
⇒ 𝜇𝜇𝑖𝑖 = 𝜇𝜇𝑖𝑖 = ⋯ = 𝜇𝜇𝑖𝑖

Chemical Engineering and Chemistry 30-1-2024 Page 53

 Non-ideal systems

• Extension for non-ideal mixtures

𝑦𝑦𝑖𝑖
Phase equilibrium ratio: 𝐾𝐾𝑖𝑖 = (K-values)
𝑥𝑥𝑖𝑖
Relative volatility between 𝐾𝐾𝑖𝑖
component i and j: 𝛼𝛼𝑖𝑖 =
𝐾𝐾𝑗𝑗

𝑝𝑝
𝜇𝜇𝑖𝑖
1
= 𝜇𝜇𝑖𝑖
2
= ⋯ = 𝜇𝜇𝑖𝑖
𝑁𝑁 𝜇𝜇𝑖𝑖 = chemical
To be calculated from: potential of comp. i
𝑝𝑝
Or equivalently: 𝑓𝑓 ̅
𝑖𝑖
1
= 𝑓𝑓 ̅ 𝑖𝑖
2
= ⋯ = 𝑓𝑓 ̅
𝑖𝑖
𝑁𝑁
𝑓𝑓𝑖𝑖̅ = partial
fugacity of comp. i
(Prausnitz et al.) 𝜇𝜇𝑖𝑖
𝑝𝑝
𝑓𝑓 ̅
𝑖𝑖 = 𝐶𝐶 exp
𝑅𝑅𝑅𝑅

Note: fugacity is more convenient because lim 𝜇𝜇𝑖𝑖 = −∞

𝑃𝑃→0

Chemical Engineering and Chemistry 30-1-2024 Page 54

 Non-ideal systems

• Extension for non-ideal mixtures

̅ = 𝑓𝑓𝑖𝑖,𝐿𝐿
VL-equilibrium: 𝑓𝑓𝑖𝑖,𝑉𝑉 ̅ (partial fugacity)

Define partial fugacity coefficient 𝜙𝜙�𝑖𝑖 as deviations to fugacity

due to pressure and composition:
̅ = 𝜙𝜙�𝑖𝑖,𝑉𝑉 𝑦𝑦𝑖𝑖 𝑃𝑃
𝑓𝑓𝑖𝑖,𝑉𝑉 and ̅ = 𝜙𝜙�𝑖𝑖,𝐿𝐿 𝑥𝑥𝑖𝑖 𝑃𝑃
𝑓𝑓𝑖𝑖,𝐿𝐿

Substitution in K-value:
̅
𝑓𝑓𝑖𝑖,𝑉𝑉 Note: for ideal mixture
𝑦𝑦𝑖𝑖 𝜙𝜙�𝑖𝑖,𝑉𝑉 𝑃𝑃 𝜙𝜙�𝑖𝑖,𝐿𝐿 𝑃𝑃𝑖𝑖𝑠𝑠
𝐾𝐾𝑖𝑖 = = ⇒ 𝐾𝐾𝑖𝑖 = 𝜙𝜙�𝑖𝑖,𝐿𝐿 =
𝑥𝑥𝑖𝑖 ̅
𝑓𝑓𝑖𝑖,𝐿𝐿 𝜙𝜙�𝑖𝑖,𝑉𝑉 𝜙𝜙�𝑖𝑖,𝑉𝑉 = 1 𝑃𝑃
𝜙𝜙�𝑖𝑖,𝐿𝐿 𝑃𝑃 ̅ = 𝑦𝑦𝑖𝑖 𝑃𝑃 𝑓𝑓𝑖𝑖,𝐿𝐿
𝑓𝑓𝑖𝑖,𝑉𝑉 ̅ = 𝑥𝑥𝑖𝑖 𝑃𝑃𝑖𝑖𝑠𝑠

Chemical Engineering and Chemistry 30-1-2024 Page 55

 Non-ideal systems

• Extension for non-ideal mixtures

Activity coefficients:
fi = pure species fugacity
𝑓𝑓𝑖𝑖
Define a pure species fugacity coefficient: 𝜙𝜙𝑖𝑖 =
𝑃𝑃
(accounting for deviations only due to pressure)
𝑓𝑓𝑖𝑖̅ 0
Define acitivity: 𝑖𝑖 𝑎𝑎 = 0 where 𝑓𝑓
𝑖𝑖 is the fugacity in a standard state
𝑓𝑓𝑖𝑖
Note: for ideal mixture
Define acitivity coefficients: 𝑎𝑎𝑖𝑖,𝑉𝑉 = 𝑦𝑦𝑖𝑖 𝑎𝑎𝑖𝑖,𝐿𝐿 = 𝑥𝑥𝑖𝑖
𝑎𝑎𝑖𝑖,𝑉𝑉 𝑎𝑎𝑖𝑖,𝐿𝐿
𝛾𝛾𝑖𝑖,𝑉𝑉 = and 𝛾𝛾𝑖𝑖,𝐿𝐿 = 𝛾𝛾𝑖𝑖,𝑉𝑉 = 1 𝛾𝛾𝑖𝑖,𝐿𝐿 = 1
𝑦𝑦𝑖𝑖 𝑥𝑥𝑖𝑖
(measure of departure from mole fractions)
𝛾𝛾𝑖𝑖,𝐿𝐿 𝜙𝜙𝑖𝑖,𝐿𝐿
̅ = 𝜙𝜙�𝑖𝑖,𝐿𝐿 𝑥𝑥𝑖𝑖 𝑃𝑃 =
⇒ 𝑓𝑓𝑖𝑖,𝐿𝐿 0
𝑎𝑎𝑖𝑖,𝐿𝐿 𝑓𝑓𝑖𝑖,𝐿𝐿 = 𝛾𝛾𝑖𝑖,𝐿𝐿 𝑥𝑥𝑖𝑖 𝜙𝜙𝑖𝑖,𝐿𝐿 𝑃𝑃 ⇒ 𝐾𝐾𝑖𝑖 = �
𝜙𝜙𝑖𝑖,𝑉𝑉
Chemical Engineering and Chemistry 30-1-2024 Page 56
 Non-ideal systems

• Extension for non-ideal mixtures

𝛾𝛾𝑖𝑖,𝐿𝐿 𝜙𝜙𝑖𝑖,𝐿𝐿
K-values: 𝐾𝐾𝑖𝑖 =
𝜙𝜙�𝑖𝑖,𝑉𝑉
Approximations:
𝑃𝑃𝑖𝑖𝑠𝑠
Raoult’s law: 𝐾𝐾𝑖𝑖 = Ideal gas + liquid
𝑃𝑃
𝛾𝛾𝑖𝑖,𝐿𝐿 𝑃𝑃𝑖𝑖𝑠𝑠 Non-ideal liquid at close to
Modified Raoult’s law: 𝐾𝐾𝑖𝑖 =
𝑃𝑃 ambient pressure
𝑠𝑠 𝑃𝑃
𝑠𝑠 𝑃𝑃𝑖𝑖 1
Poynting correction: 𝐾𝐾𝑖𝑖 = 𝛾𝛾𝑖𝑖,𝐿𝐿 𝜙𝜙𝑖𝑖,𝑉𝑉 exp � 𝑣𝑣 𝑑𝑑𝑑𝑑
𝑃𝑃 𝑅𝑅𝑅𝑅 𝑃𝑃𝑠𝑠 𝑖𝑖,𝐿𝐿
𝑖𝑖
𝐻𝐻𝑖𝑖
Henry’s law: 𝐾𝐾𝑖𝑖 = For species at supercritical
𝑃𝑃 temperature

Chemical Engineering and Chemistry 30-1-2024 Page 57

 Non-ideal systems

• Extension for non-ideal mixtures

Non-ideal models:
EOS-models (equation-of-state): 𝑃𝑃 = 𝑃𝑃(𝑇𝑇, 𝑣𝑣)
𝑍𝑍𝑍𝑍𝑍𝑍
− Generalized Ideal gas: 𝑃𝑃 =
𝑣𝑣 accounting for
excluded volume,
𝑅𝑅𝑅𝑅 𝑎𝑎 differences in molar
− Soave-Redlich-Kwong: 𝑃𝑃 = − 2
𝑣𝑣 − 𝑏𝑏 𝑣𝑣 + 𝑏𝑏𝑏𝑏 shape and molecular
interactions
𝑅𝑅𝑅𝑅 𝑎𝑎
− Peng-Robinson: 𝑃𝑃 = −
𝑣𝑣 − 𝑏𝑏 𝑣𝑣 2 + 2𝑏𝑏𝑏𝑏 − 𝑏𝑏 2

Activity coefficient/free energy models:

Margules, Van Laar, Wilson, NRTL, UNIQUAC, UNIFAC

Chemical Engineering and Chemistry 30-1-2024 Page 58

 Equilibrium-based models

• The Equilibrium Stage Model

𝐿𝐿𝑗𝑗−1 𝑉𝑉𝑗𝑗
Liquid from 𝑥𝑥𝑖𝑖,𝑗𝑗−1 𝑦𝑦𝑖𝑖,𝑗𝑗
stage above ℎ𝐿𝐿,𝑗𝑗−1 ℎ𝑉𝑉,𝑗𝑗 Vapor
𝑊𝑊𝑗𝑗 side stream
𝑇𝑇𝑗𝑗−1 𝑇𝑇𝑗𝑗
𝑧𝑧𝑖𝑖,𝑗𝑗 ℎ𝐹𝐹,𝑗𝑗 𝑃𝑃𝑗𝑗−1 𝑃𝑃𝑗𝑗
𝑇𝑇𝐹𝐹,𝑗𝑗 𝑃𝑃𝐹𝐹,𝑗𝑗
Feed 𝐹𝐹𝑗𝑗 Equilibrium stage j 𝑄𝑄𝑗𝑗 Heat
transfer
𝑥𝑥𝑖𝑖,𝑗𝑗 𝑦𝑦𝑖𝑖,𝑗𝑗+1
Liquid 𝑈𝑈 ℎ𝐿𝐿,𝑗𝑗 ℎ𝑉𝑉,𝑗𝑗+1
𝑗𝑗
side stream 𝑇𝑇𝑗𝑗 𝑇𝑇𝑗𝑗+1
Vapor from
𝑃𝑃𝑗𝑗 𝑃𝑃𝑗𝑗+1 stage below
𝐿𝐿𝑗𝑗 𝑉𝑉𝑗𝑗+1

Chemical Engineering and Chemistry 30-1-2024 Page 59

 Equilibrium-based models

• The Equilibrium Stage Model

MESH equations:
M equations: Material balances for each component
𝑀𝑀𝑖𝑖,𝑗𝑗 = 𝐿𝐿𝑗𝑗−1 𝑥𝑥𝑖𝑖,𝑗𝑗−1 + 𝑉𝑉𝑗𝑗+1 𝑦𝑦𝑖𝑖,𝑗𝑗+1 + 𝐹𝐹𝑗𝑗 𝑧𝑧𝑖𝑖,𝑗𝑗 − 𝐿𝐿𝑗𝑗 + 𝑈𝑈𝑗𝑗 𝑥𝑥𝑖𝑖,𝑗𝑗 − 𝑉𝑉𝑗𝑗 + 𝑊𝑊𝑗𝑗 𝑦𝑦𝑖𝑖,𝑗𝑗 = 0

E equations: phase-equilibrium relations for each component

𝑀𝑀𝑀𝑀 𝑀𝑀𝑀𝑀
𝐸𝐸𝑖𝑖,𝑗𝑗 = 𝐸𝐸𝑖𝑖,𝑗𝑗 𝐾𝐾𝑖𝑖,𝑗𝑗 𝑥𝑥𝑖𝑖,𝑗𝑗 − 𝑦𝑦𝑖𝑖,𝑗𝑗 + 1 − 𝐸𝐸𝑖𝑖,𝑗𝑗 𝑦𝑦𝑖𝑖,𝑗𝑗+1 = 0

S equations: mole-fraction summation equations

𝑆𝑆𝑦𝑦 𝐶𝐶
= ∑𝑖𝑖=1 𝑦𝑦𝑖𝑖,𝑗𝑗 − 1 = 0 𝑆𝑆𝑥𝑥 𝑗𝑗 = ∑𝐶𝐶𝑖𝑖=1 𝑥𝑥𝑖𝑖,𝑗𝑗 − 1 = 0
𝑗𝑗

H equation: energy balance

𝐻𝐻𝑗𝑗 = 𝐿𝐿𝑗𝑗−1 ℎ𝐿𝐿,𝑗𝑗−1 + 𝑉𝑉𝑗𝑗+1 ℎ𝑉𝑉,𝑗𝑗+1 + 𝐹𝐹𝑗𝑗 ℎ𝐹𝐹,𝑗𝑗
− 𝐿𝐿𝑗𝑗 + 𝑈𝑈𝑗𝑗 ℎ𝐿𝐿,𝑗𝑗 − 𝑉𝑉𝑗𝑗 + 𝑊𝑊𝑗𝑗 ℎ𝑉𝑉,𝑗𝑗 − 𝑄𝑄𝑗𝑗 = 0
 Equilibrium-based models

• The Equilibrium Stage Model

MESH equations:
Number of equations: 𝑁𝑁 2𝐶𝐶 + 3 (non-linear algebraic eqns)

Number of unknowns: 𝑥𝑥𝑖𝑖,𝑗𝑗 𝑦𝑦𝑖𝑖,𝑗𝑗 𝐿𝐿𝑗𝑗 𝑉𝑉𝑗𝑗 𝑇𝑇𝑗𝑗 𝑗𝑗 = 1 … 𝑁𝑁

⇒ Wide variety of numerical solution procedures available:

1930’s: stage-by-stage computations
(only stable for simplified thermodynamics)
>1960’s: different optimal techniques for different problems:
• Equation tearing procedures (tear variables Tj, Vj):
Tridiagonal matrix algorithm, Bubble-point method,
Sum-rates method
• Newton-Raphson method
Chemical Engineering and Chemistry 30-1-2024 Page 61
 Equilibrium-based models

• Tray efficiencies determined from mass transfer model

Simplified assumptions required

to model gas-liquid contacting
Mass transfer rate:
• Driving force: gradient in concentration/activity
• Exchange surface area (bubbles)
6𝜖𝜖𝑣𝑣
(tray column: 30 < a < 100 m2/m3) 𝑎𝑎 =
𝑑𝑑
Chemical Engineering and Chemistry 30-1-2024 Page 62
Equilibrium-based models
• Tray mass transfer model: film model
− Thin film layer on both sides of the interface
− In the thin film only diffusion occurs
− Simplified (linear) concentration profiles in de film
− No gradients in the bulk
𝑑𝑑𝑁𝑁
𝑁𝑁𝑖𝑖,𝑉𝑉 = 𝐽𝐽𝑖𝑖,𝑉𝑉 + 𝑦𝑦𝑖𝑖 𝑁𝑁𝑡𝑡𝑡𝑡𝑡𝑡 In case of equimolar flow: 𝑁𝑁𝑡𝑡𝑡𝑡𝑡𝑡 = 0 & 𝑖𝑖 = 0
𝑑𝑑𝑑𝑑
𝑑𝑑𝑦𝑦 𝑖𝑖 𝑑𝑑𝑥𝑥 𝑖𝑖
𝑁𝑁𝑖𝑖,𝐿𝐿 = 𝐽𝐽𝑖𝑖,𝐿𝐿 + 𝑥𝑥𝑖𝑖 𝑁𝑁𝑡𝑡𝑡𝑡𝑡𝑡 𝐽𝐽𝑖𝑖,𝑉𝑉 𝑉𝑉
= −𝐷𝐷𝑖𝑖,𝑉𝑉 𝑐𝑐𝑡𝑡𝑡𝑡𝑡𝑡 𝐿𝐿
= −𝐷𝐷𝑖𝑖,𝐿𝐿 𝑐𝑐𝑡𝑡𝑡𝑡𝑡𝑡 = 𝐽𝐽𝑖𝑖,𝐿𝐿
𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑

m/s

m/s

30-1-2024
 Equilibrium-based models

• Tray mass transfer model: film model

− Gas phase flux = liquid phase flux:
𝑉𝑉
𝐽𝐽𝑖𝑖 = 𝑘𝑘𝐺𝐺 𝑐𝑐𝑡𝑡𝑡𝑡𝑡𝑡 𝑦𝑦𝑖𝑖,𝑏𝑏 − 𝑦𝑦𝑖𝑖𝐼𝐼
𝐿𝐿
= 𝑘𝑘𝐿𝐿 𝑐𝑐𝑡𝑡𝑡𝑡𝑡𝑡 𝑥𝑥𝑖𝑖𝐼𝐼 − 𝑥𝑥𝑖𝑖,𝑏𝑏
𝑦𝑦𝑖𝑖𝐼𝐼
Elimination of interface fractions with: 𝐾𝐾𝑖𝑖 =
𝑥𝑥𝑖𝑖𝐼𝐼
1
𝑉𝑉 𝐽𝐽𝑖𝑖 = 𝑦𝑦𝑖𝑖,𝑏𝑏 − 𝑦𝑦𝑖𝑖𝐼𝐼 𝑉𝑉
𝐽𝐽𝑖𝑖 = 𝐾𝐾𝐺𝐺 𝑐𝑐𝑡𝑡𝑡𝑡𝑡𝑡 𝑦𝑦𝑖𝑖,𝑏𝑏 − 𝑦𝑦𝑖𝑖∗
𝑘𝑘𝐺𝐺 𝑐𝑐𝑡𝑡𝑡𝑡𝑡𝑡
⇒
1 𝑦𝑦𝑖𝑖𝐼𝐼 1 1 𝑉𝑉
𝐾𝐾𝑖𝑖 𝑐𝑐𝑡𝑡𝑡𝑡𝑡𝑡
𝐿𝐿 𝐽𝐽𝑖𝑖
𝑘𝑘𝐿𝐿 𝑐𝑐𝑡𝑡𝑡𝑡𝑡𝑡
=
𝐾𝐾𝑖𝑖
− 𝑥𝑥𝑖𝑖,𝑏𝑏 = + 𝐿𝐿 𝑦𝑦𝑖𝑖∗ = 𝐾𝐾𝑖𝑖 𝑥𝑥𝑖𝑖,𝑏𝑏
𝐾𝐾𝐺𝐺 𝑘𝑘𝐺𝐺 𝑘𝑘𝐿𝐿 𝑐𝑐𝑡𝑡𝑡𝑡𝑡𝑡

Distillation/stripping: Reactive absorption: Absorption:

kG >> kL kG << kLEA kG ~ kL
 Equilibrium-based models

• Tray mass transfer model: film model

− Effects of non-equimolar counter diffusion: drift-flux
(or bulk-flow effect)

Consider binary case with NB=0 (inert) ⇒ 𝑁𝑁𝑡𝑡𝑡𝑡𝑡𝑡 = 𝑁𝑁𝐴𝐴

(e.g. evaporation of a component from a liquid under N2 atmosphere)

𝑁𝑁𝐴𝐴,𝑉𝑉 = 𝐽𝐽𝐴𝐴,𝑉𝑉 + 𝑦𝑦𝐴𝐴 𝑁𝑁𝑡𝑡𝑡𝑡𝑡𝑡 = 𝐽𝐽𝐴𝐴,𝑉𝑉 + 𝑦𝑦𝐴𝐴 𝑁𝑁𝐴𝐴

𝑁𝑁𝐵𝐵,𝑉𝑉 = 𝐽𝐽𝐵𝐵,𝑉𝑉 + 𝑦𝑦𝐵𝐵 𝑁𝑁𝑡𝑡𝑡𝑡𝑡𝑡 = 𝐽𝐽𝐵𝐵,𝑉𝑉 + 𝑦𝑦𝐵𝐵 𝑁𝑁𝐴𝐴 = 0
(i.e. diffusion and drift-flux cancel each other out)
𝐽𝐽𝐴𝐴,𝑉𝑉 𝑉𝑉
1 𝑑𝑑𝑦𝑦𝐴𝐴
⇒ 𝑁𝑁𝐴𝐴,𝑉𝑉 = = −𝐷𝐷𝑖𝑖,𝑉𝑉 𝑐𝑐𝑡𝑡𝑡𝑡𝑡𝑡 = 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐
1 − 𝑦𝑦𝐴𝐴 1 − 𝑦𝑦𝐴𝐴 𝑑𝑑𝑑𝑑

Solve with: 𝑧𝑧 = 0: 𝑦𝑦𝐴𝐴 = 𝑦𝑦𝐴𝐴,𝑏𝑏 and 𝑧𝑧 = 𝛿𝛿: 𝑦𝑦𝐴𝐴 = 𝑦𝑦𝐴𝐴𝐼𝐼

Chemical Engineering and Chemistry 30-1-2024 Page 65

 Equilibrium-based models

• Tray mass transfer model: film model

− Effects of non-equimolar counter diffusion: drift-flux
(or bulk-flow effect)

Solution:
𝑧𝑧
1 − 𝑦𝑦𝐴𝐴 (𝑧𝑧) 1− 𝑦𝑦𝐴𝐴𝐼𝐼 𝛿𝛿 𝐷𝐷𝑖𝑖,𝑉𝑉 𝑉𝑉 1 − 𝑦𝑦𝐴𝐴𝐼𝐼
= 𝑁𝑁𝐴𝐴,𝑉𝑉 = 𝑐𝑐 ln
1 − 𝑦𝑦𝐴𝐴,𝑏𝑏 1 − 𝑦𝑦𝐴𝐴,𝑏𝑏 𝛿𝛿 𝑡𝑡𝑡𝑡𝑡𝑡 1 − 𝑦𝑦𝐴𝐴,𝑏𝑏

𝐷𝐷𝑖𝑖,𝑉𝑉 𝑉𝑉 1
Or rewritten: 𝑁𝑁𝐴𝐴,𝑉𝑉 = 𝑐𝑐𝑡𝑡𝑡𝑡𝑡𝑡 𝑦𝑦𝐴𝐴,𝑏𝑏 − 𝑦𝑦𝐴𝐴𝐼𝐼
𝛿𝛿 1 − 𝑦𝑦𝐴𝐴 𝐿𝐿𝐿𝐿

𝑦𝑦𝐴𝐴,𝑏𝑏 − 𝑦𝑦𝐴𝐴𝐼𝐼
with 1 − 𝑦𝑦𝐴𝐴 𝐿𝐿𝐿𝐿 =
ln 1 − 𝑦𝑦𝐴𝐴𝐼𝐼 � 1 − 𝑦𝑦𝐴𝐴,𝑏𝑏

Chemical Engineering and Chemistry 30-1-2024 Page 66

 Equilibrium-based models

• Tray mass transfer model: Flow pattern on a tray

Murphree model (oldest, but still most widely used):
− Uniform liquid composition on the tray equal to that leaving the
tray (i.e. well mixed liquid bulk)
− Plug flow of the vapor passing up through the liquid

Chemical Engineering and Chemistry 30-1-2024

30-1-2024 Page 67
 Equilibrium-based models

• Tray mass transfer model: Murphree model

Differential component material balance in the gas phase:
𝑑𝑑𝑑𝑑𝑖𝑖 𝑎𝑎 = specific interfacial area
𝑉𝑉
= −𝐾𝐾𝐺𝐺 𝑐𝑐𝑡𝑡𝑡𝑡𝑡𝑡 𝑎𝑎 𝑦𝑦𝑖𝑖 − 𝑦𝑦𝑖𝑖∗ 𝐴𝐴𝑏𝑏
𝑑𝑑𝑑𝑑 𝐴𝐴𝑏𝑏 = active bubbling area
Using: 𝑛𝑛𝑖𝑖 = 𝑦𝑦𝑖𝑖 𝑉𝑉
𝑉𝑉 = molar gas flow rate
𝑑𝑑𝑑𝑑
Assuming: ≈0
𝑑𝑑𝑑𝑑
𝑑𝑑𝑑𝑑𝑖𝑖
⇒ 𝑉𝑉 𝑉𝑉
= −𝐾𝐾𝐺𝐺 𝑐𝑐𝑡𝑡𝑡𝑡𝑡𝑡 𝑎𝑎 𝑦𝑦𝑖𝑖 − 𝑦𝑦𝑖𝑖∗ 𝐴𝐴𝑏𝑏
𝑑𝑑𝑑𝑑

Integration:
𝑦𝑦𝑖𝑖,𝑛𝑛 ℎ𝑓𝑓 𝑉𝑉
𝑑𝑑𝑑𝑑𝑖𝑖 𝐾𝐾𝐺𝐺 𝑐𝑐𝑡𝑡𝑡𝑡𝑡𝑡 𝑎𝑎𝐴𝐴𝑏𝑏
� ∗
= −� 𝑑𝑑𝑑𝑑
𝑦𝑦𝑖𝑖,𝑛𝑛+1 𝑦𝑦𝑖𝑖 − 𝑦𝑦𝑖𝑖,𝑛𝑛 0 𝑉𝑉

Chemical Engineering and Chemistry 30-1-2024

30-1-2024 Page 68
 Equilibrium-based models

• Tray mass transfer model: Murphree model

𝑦𝑦𝑖𝑖,𝑛𝑛 ℎ𝑓𝑓 𝑉𝑉
𝑑𝑑𝑑𝑑𝑖𝑖 𝐾𝐾𝐺𝐺 𝑐𝑐𝑡𝑡𝑡𝑡𝑡𝑡 𝑎𝑎𝐴𝐴𝑏𝑏
Integration: � ∗
=� 𝑑𝑑𝑑𝑑
𝑦𝑦𝑖𝑖,𝑛𝑛+1 𝑦𝑦𝑖𝑖 − 𝑦𝑦𝑖𝑖,𝑛𝑛 0 𝑉𝑉

∗ 𝑉𝑉
𝑦𝑦𝑖𝑖,𝑛𝑛 − 𝑦𝑦𝑖𝑖,𝑛𝑛 𝐾𝐾𝐺𝐺 𝑐𝑐𝑡𝑡𝑡𝑡𝑡𝑡 𝑎𝑎𝐴𝐴𝑏𝑏 ℎ𝑓𝑓
⇒ ln ∗ = −𝑁𝑁𝑂𝑂𝑂𝑂 𝑁𝑁𝑂𝑂𝑂𝑂 =
𝑦𝑦𝑖𝑖,𝑛𝑛+1 − 𝑦𝑦𝑖𝑖,𝑛𝑛 𝑉𝑉

Rearrangement in terms of approach to equilibrium:

∗
𝑦𝑦𝑖𝑖,𝑛𝑛 − 𝑦𝑦𝑖𝑖,𝑛𝑛 𝑦𝑦𝑖𝑖,𝑛𝑛+1 − 𝑦𝑦𝑖𝑖,𝑛𝑛
1− Note that EMV,i
∗ = ∗ = 1 − exp −𝑁𝑁𝑂𝑂𝑂𝑂
𝑦𝑦𝑖𝑖,𝑛𝑛+1 − 𝑦𝑦𝑖𝑖,𝑛𝑛 𝑦𝑦𝑖𝑖,𝑛𝑛+1 − 𝑦𝑦𝑖𝑖,𝑛𝑛 can be negative!

⇒ Murphree vapor efficiency: 𝑉𝑉

𝑦𝑦𝑖𝑖,𝑛𝑛+1 − 𝑦𝑦𝑖𝑖,𝑛𝑛 𝐾𝐾𝐺𝐺 𝑐𝑐𝑡𝑡𝑡𝑡𝑡𝑡 𝑎𝑎𝐴𝐴𝑏𝑏 ℎ𝑓𝑓
𝐸𝐸𝑀𝑀𝑀𝑀,𝑖𝑖 = ∗ = 1 − exp −𝑁𝑁𝑂𝑂𝑂𝑂 𝑁𝑁𝑂𝑂𝑂𝑂 =
𝑦𝑦𝑖𝑖,𝑛𝑛+1 − 𝑦𝑦𝑖𝑖,𝑛𝑛 𝑉𝑉
Chemical Engineering and Chemistry 30-1-2024
30-1-2024 Page 69
 Equilibrium-based models

• Overall column efficiency:

Use Kremser’s equation
(in case of constant S):

𝑁𝑁𝑒𝑒𝑒𝑒 ln 1 + 𝐸𝐸𝑀𝑀𝑀𝑀 𝑆𝑆 − 1
𝐸𝐸𝑜𝑜𝑜𝑜 = =
𝑁𝑁 ln 𝑆𝑆

𝐸𝐸𝑀𝑀𝑀𝑀 = 1 − exp −𝑁𝑁𝑂𝑂𝑂𝑂

𝑉𝑉
𝐾𝐾𝐺𝐺 𝑐𝑐𝑡𝑡𝑡𝑡𝑡𝑡 𝑎𝑎𝐴𝐴𝑏𝑏 ℎ𝑓𝑓
𝑁𝑁𝑂𝑂𝑂𝑂 = S EO
𝑉𝑉 1.001 0.700
Given: 1.01 0.701
Separation factor S:
EMV = 0.70 1.1 0.710
• S = KV/L (distillation, stripping) 1.5 0.740
Eov = ?
• S = L/KV (absorption) 2 0.766
5 0.829
30-1-2024

Chemical Engineering and Chemistry 30-1-2024

30-1-2024 Page 70
 Rate-based models

• Problems with Equilibrium-based models

− Assumption of equilibrium of exiting phase composition is
usually not valid;
− Introduction of overall stage efficiencies, but these vary over a
range of 5 – 120%.
− Introduction of Murphree vapor tray efficiency 𝐸𝐸𝑀𝑀𝑀𝑀 𝑖𝑖,𝑗𝑗
• Requires assumptions to describe mass transfer on the tray:
- uniform concentration in streams
- complete liquid phase mixing
- plug flow behavior of vapor
• How to deal with multi-component interaction effects?
For multicomponent mixtures 𝐸𝐸𝑀𝑀𝑀𝑀 𝑖𝑖,𝑗𝑗 varies per component
and per tray (can even be negative!), and due to interaction
effects from −∞ to ∞
Chemical Engineering and Chemistry 30-1-2024 Page 71
 Rate-based models

• The Non-equilibrium Stage Model

𝐿𝐿𝑗𝑗−1 𝑉𝑉𝑗𝑗
Liquid from 𝑥𝑥𝑖𝑖,𝑗𝑗−1 𝑦𝑦𝑖𝑖,𝑗𝑗 𝑟𝑟𝑗𝑗𝑉𝑉
stage above ℎ𝐿𝐿,𝑗𝑗−1 ℎ𝑉𝑉,𝑗𝑗 Vapor
𝑊𝑊𝑗𝑗 side stream
𝐿𝐿
𝐿𝐿
𝑧𝑧𝑖𝑖,𝑗𝑗 𝐿𝐿
ℎ𝐹𝐹,𝑗𝑗 𝑇𝑇𝑗𝑗−1 𝑇𝑇𝑗𝑗𝑉𝑉 𝐿𝐿 𝐿𝐿
𝑧𝑧𝑖𝑖,𝑗𝑗 ℎ𝐹𝐹,𝑗𝑗
𝐿𝐿
𝑇𝑇𝐹𝐹,𝑗𝑗 𝐿𝐿
𝑃𝑃𝐹𝐹,𝑗𝑗 𝑃𝑃𝑗𝑗−1 𝑃𝑃𝑗𝑗 𝐿𝐿 𝐿𝐿
𝑇𝑇𝐹𝐹,𝑗𝑗 𝑃𝑃𝐹𝐹,𝑗𝑗
Liquid 𝐹𝐹 𝐿𝐿 stage j 𝐹𝐹𝑗𝑗𝑉𝑉 Vapor
Feed 𝑗𝑗 𝑁𝑁𝑖𝑖 Feed
𝑄𝑄𝑗𝑗𝐿𝐿 𝐸𝐸 𝑄𝑄𝑗𝑗𝑉𝑉
𝑥𝑥𝑖𝑖,𝑗𝑗 𝑦𝑦𝑖𝑖,𝑗𝑗+1
Liquid 𝑈𝑈 ℎ𝐿𝐿,𝑗𝑗 ℎ𝑉𝑉,𝑗𝑗+1
𝑗𝑗
side stream 𝐿𝐿 𝑇𝑇𝑗𝑗𝐿𝐿 𝑉𝑉
𝑇𝑇𝑗𝑗+1
𝑟𝑟𝑗𝑗 Vapor from
𝑃𝑃𝑗𝑗 𝑃𝑃𝑗𝑗+1 stage below
𝐿𝐿𝑗𝑗 𝑉𝑉𝑗𝑗+1

Chemical Engineering and Chemistry 30-1-2024 Page 72

 Rate-based models

• The Non-equilibrium Stage Model

− Separate material balances for liquid and vapor phases around
each stage
− Stage can be a tray, a collection of trays, or a segment of a
packed column
− Sherwood and Nusselt correlations required for the mass and
heat transport
− Account for multi-component transport including effects of drift-
fluxes because of a net mass flux across the interface and non-
idealities (via (linearized) Maxwell-Stefan equations)

Chemical Engineering and Chemistry 30-1-2024 Page 73

 Rate-based models

• The Non-equilibrium Stage Model

MERSHQ equations:
Liquid-phase component material balances:
𝐿𝐿
𝑀𝑀𝑖𝑖,𝑗𝑗 = 1 + 𝑟𝑟𝑗𝑗𝐿𝐿 𝐿𝐿𝑗𝑗 𝑥𝑥𝑖𝑖,𝑗𝑗 − 𝐿𝐿𝑗𝑗−1 𝑥𝑥𝑖𝑖,𝑗𝑗−1 − 𝑧𝑧𝑖𝑖,𝑗𝑗
𝐿𝐿 𝐿𝐿 𝐿𝐿
𝐹𝐹𝑗𝑗 − 𝑁𝑁𝑖𝑖,𝑗𝑗 =0 i = 1, 2, …, C

Vapor-phase component material balances:

𝑉𝑉
𝑀𝑀𝑖𝑖,𝑗𝑗 = 1 + 𝑟𝑟𝑗𝑗𝑉𝑉 𝑉𝑉𝑗𝑗 𝑦𝑦𝑖𝑖,𝑗𝑗 − 𝑉𝑉𝑗𝑗+1 𝑦𝑦𝑖𝑖,𝑗𝑗+1 − 𝑧𝑧𝑖𝑖,𝑗𝑗
𝑉𝑉 𝑉𝑉 𝑉𝑉
𝐹𝐹𝑗𝑗 − 𝑁𝑁𝑖𝑖,𝑗𝑗 =0 i = 1, 2, …, C

Liquid-phase energy balance:

𝐸𝐸𝑗𝑗𝐿𝐿 = 1 + 𝑟𝑟𝑗𝑗𝐿𝐿 𝐿𝐿𝑗𝑗 𝐻𝐻𝑗𝑗𝐿𝐿 − 𝐿𝐿𝑗𝑗−1 𝐻𝐻𝑗𝑗−1
𝐿𝐿
− 𝐹𝐹𝑗𝑗𝐿𝐿 𝐻𝐻𝑗𝑗𝐹𝐹,𝐿𝐿 + 𝑄𝑄𝑗𝑗𝐿𝐿 − 𝑒𝑒𝑗𝑗𝐿𝐿 = 0
Vapor-phase energy balance:
𝐸𝐸𝑗𝑗𝑉𝑉 = 1 + 𝑟𝑟𝑗𝑗𝑉𝑉 𝑉𝑉𝑗𝑗 𝐻𝐻𝑗𝑗𝑉𝑉 − 𝑉𝑉𝑗𝑗+1 𝐻𝐻𝑗𝑗+1
𝑉𝑉
− 𝐹𝐹𝑗𝑗𝑉𝑉 𝐻𝐻𝑗𝑗𝐹𝐹,𝑉𝑉 + 𝑄𝑄𝑗𝑗𝑉𝑉 − 𝑒𝑒𝑗𝑗𝑉𝑉 = 0
Continuity of energy and mass transfer rates across the VL-interface:
𝐸𝐸𝑗𝑗𝐼𝐼 = 𝑒𝑒𝑗𝑗𝑉𝑉 − 𝑒𝑒𝑗𝑗𝐿𝐿 = 0 𝐿𝐿
𝑅𝑅𝑖𝑖,𝑗𝑗 𝐿𝐿
= 𝑁𝑁𝑖𝑖,𝑗𝑗 − 𝑁𝑁𝑖𝑖,𝑗𝑗 =0 i = 1, 2, …, C-1
𝑉𝑉 𝑉𝑉
𝑅𝑅𝑖𝑖,𝑗𝑗 = 𝑁𝑁𝑖𝑖,𝑗𝑗 − 𝑁𝑁𝑖𝑖,𝑗𝑗 =0 i = 1, 2, …, C-1
 Rate-based models

• The Non-equilibrium Stage Model

Summation equations:
𝑆𝑆𝑗𝑗𝐿𝐿,𝐼𝐼 = ∑𝑖𝑖=1
𝐶𝐶 𝐼𝐼
𝑥𝑥𝑖𝑖,𝑗𝑗 −1=0
𝑆𝑆𝑗𝑗𝑉𝑉,𝐼𝐼 = ∑𝐶𝐶𝑖𝑖=1 𝑦𝑦𝑖𝑖,𝑗𝑗
𝐼𝐼
−1=0

Hydraulic equation for stage pressure drop:

𝐻𝐻𝑗𝑗 = 𝑃𝑃𝑗𝑗+1 −𝑃𝑃𝑗𝑗 − Δ𝑃𝑃𝑗𝑗 = 0
Stage is at mechanical equilibrium: 𝑃𝑃𝑗𝑗 = 𝑃𝑃𝑗𝑗𝐿𝐿 = 𝑃𝑃𝑗𝑗𝑉𝑉

Phase equilibrium at interface:

𝐼𝐼 𝐼𝐼 𝐼𝐼
𝑄𝑄𝑖𝑖,𝑗𝑗 = 𝐾𝐾𝑖𝑖,𝑗𝑗 𝑥𝑥𝑖𝑖,𝑗𝑗 −𝑦𝑦𝑖𝑖,𝑗𝑗 =0 i = 1, 2, …, C

Chemical Engineering and Chemistry 30-1-2024 Page 75

 Rate-based models

• Transport rate equations

Mass transfer (diffusion + drift)
𝑉𝑉 𝑉𝑉
𝑁𝑁𝑖𝑖,𝑗𝑗 = 𝑎𝑎𝑗𝑗𝐼𝐼 𝐽𝐽𝑖𝑖,𝑗𝑗 + 𝑦𝑦𝑖𝑖,𝑗𝑗 𝑁𝑁𝑡𝑡𝑡𝑡𝑡𝑡,𝑗𝑗
𝐽𝐽𝑖𝑖,𝑗𝑗 = Molar diffusion flux relative to
𝐿𝐿/𝑉𝑉
𝐿𝐿 the molar average velocity
𝑁𝑁𝑖𝑖,𝑗𝑗 = 𝑎𝑎𝑗𝑗𝐼𝐼 𝐽𝐽𝑖𝑖,𝑗𝑗
𝐿𝐿
+ 𝑥𝑥𝑖𝑖,𝑗𝑗 𝑁𝑁𝑡𝑡𝑡𝑡𝑡𝑡,𝑗𝑗 𝐶𝐶 𝐿𝐿/𝑉𝑉
∑𝑖𝑖=1 𝐽𝐽𝑖𝑖,𝑗𝑗 = 0

Energy transfer
𝐶𝐶
𝑉𝑉 � 𝑉𝑉
𝑒𝑒𝑗𝑗𝑉𝑉 = 𝑎𝑎𝑗𝑗𝐼𝐼 ℎ𝑉𝑉 𝑇𝑇 𝑉𝑉 − 𝑇𝑇 𝐼𝐼 + � 𝑁𝑁𝑖𝑖,𝑗𝑗 𝐻𝐻𝑖𝑖,𝑗𝑗
𝑖𝑖=1
𝐶𝐶 “thermal drift”
𝐿𝐿 � 𝐿𝐿
𝑒𝑒𝑗𝑗𝐿𝐿 = 𝑎𝑎𝑗𝑗𝐼𝐼 ℎ𝐿𝐿 𝑇𝑇 𝐼𝐼 − 𝑇𝑇 𝐿𝐿 + � 𝑁𝑁𝑖𝑖,𝑗𝑗 𝐻𝐻𝑖𝑖,𝑗𝑗
𝑖𝑖=1

Chemical Engineering and Chemistry 30-1-2024 Page 76

 Rate-based models

• Mass transfer coefficients

For a multi-component mixture: For a binary mixture:
𝑉𝑉
𝑱𝑱𝑉𝑉𝑗𝑗 = 𝑐𝑐𝑡𝑡𝑡𝑡𝑡𝑡
𝑉𝑉
𝜿𝜿𝑉𝑉 𝒚𝒚𝑉𝑉 − 𝒚𝒚𝐼𝐼 𝑎𝑎𝑎𝑎𝑎𝑎 𝐽𝐽𝑖𝑖,𝑗𝑗 𝑉𝑉
= 𝑐𝑐𝑡𝑡𝑡𝑡𝑡𝑡 𝑘𝑘𝑖𝑖𝑉𝑉 𝑦𝑦𝑖𝑖𝑉𝑉 − 𝑦𝑦𝑖𝑖𝐼𝐼 𝑎𝑎𝑎𝑎𝑎𝑎
𝑱𝑱𝐿𝐿𝑗𝑗 = 𝑐𝑐𝑡𝑡𝑡𝑡𝑡𝑡
𝐿𝐿
𝜿𝜿𝐿𝐿 𝒙𝒙𝑉𝑉 − 𝒙𝒙𝐼𝐼 𝑎𝑎𝑎𝑎𝑎𝑎
𝐿𝐿
𝐽𝐽𝑖𝑖,𝑗𝑗 𝐿𝐿
= 𝑐𝑐𝑡𝑡𝑡𝑡𝑡𝑡 𝑘𝑘𝑖𝑖𝐿𝐿 𝑥𝑥𝑖𝑖𝑉𝑉 − 𝑥𝑥𝑖𝑖𝐼𝐼 𝑎𝑎𝑎𝑎𝑎𝑎
(vectors for C - 1 components)

𝜿𝜿𝑃𝑃 = 𝜿𝜿𝑃𝑃0 𝝃𝝃𝑃𝑃 flux correction factors + Sh and Nu

𝜿𝜿𝑃𝑃0 = 𝑹𝑹𝑃𝑃 −1 𝚪𝚪 𝑃𝑃 Non-idealities correlations

𝑃𝑃 1 1 𝑧𝑧𝑖𝑖 𝑧𝑧𝑘𝑘
𝑅𝑅𝑖𝑖,𝑗𝑗 = −𝑧𝑧𝑖𝑖 𝑃𝑃
𝑘𝑘𝑖𝑖𝑖𝑖
− 𝑃𝑃
𝑘𝑘𝑖𝑖𝑖𝑖
𝑃𝑃
𝑅𝑅𝑖𝑖,𝑖𝑖 = 𝑃𝑃 + ∑𝐶𝐶𝑘𝑘=1 𝑃𝑃
𝑘𝑘𝑖𝑖𝑖𝑖 𝑘𝑘𝑖𝑖𝑖𝑖
𝑘𝑘≠𝑖𝑖
𝑃𝑃 𝜕𝜕 ln 𝛾𝛾𝑖𝑖
Γ𝑖𝑖,𝑗𝑗 = 𝛿𝛿𝑖𝑖𝑖𝑖 + 𝑥𝑥𝑖𝑖 (z = x or y; P = L or V)
𝜕𝜕𝑥𝑥𝑗𝑗
𝑇𝑇,𝑃𝑃,𝑥𝑥𝑘𝑘 ,𝑘𝑘≠𝑗𝑗

Chemical Engineering and Chemistry 30-1-2024 Page 77

 Rate-based models

• Developoment of rate-based models:

− Rate based models available in process simulators:
− ChemSep (R. Taylor)
− RATEFRAC in ASPEN PLUS,
− CHEMCAD
− …

Chemical Engineering and Chemistry 30-1-2024 Page 78

 Conclusions

• Start with simple short-cut methods to gain a general

idea of the separation. This will also provide useful
starting guesses for more rigorous calculations

• For multi-component mixtures with large non-

idealities use rate-based models, available in
process simulators. Check careful the used
thermodynamics, mass/heat transfer correlations,
etc.

• For intensified, integrated or reactive separations,

you may need to develop your own program…

Chemical Engineering and Chemistry 30-1-2024 Page 79