Chapter 1

Continuous rectification of binary mixtures

1. Introduction
Unit operations are methods of separation of constituents and purification (liquid-liquid
extraction, absorption, adsorption, distillation…etc). Rectification (distillation with reflux) is the
technique of separating the volatile constituents of a mixture in order to obtain them in an almost
pure state. It is considered among the main operations in the chemical and petrochemical
industries and especially in refineries.

2. The distillation column

2.1 Distillation plant

𝐂𝐨𝐧𝐝𝐞𝐧𝐬𝐞𝐫
𝐇𝐞𝐚𝐭 𝐄𝐱𝐜𝐡𝐚𝐧𝐠𝐞𝐫

𝐃, 𝐱𝐃
Distillate (liquid)

𝐅, 𝐳𝐅
Feed

𝐑𝐞𝐛𝐨𝐢𝐥𝐞𝐫
Heat Exchanger
𝐁, 𝐱𝐁
Bottom (liquid)

Figure 1 : Distillation plant

The distillation unit is continuously fed by a load F and provides a distillate D and a residue
B, it is composed of three main parts :
 A contact system.
 A steam system (Boiler).
 A condensation system (Condenser).
2.2 Operating mode

- Most distillation columns in industry operate continuously. They can only be stopped for
servicing or in the event of a breakdown.
- Generally, only one mixture is processed in a single installation.
- The load that permanently feeds the column is characterized by a composition, flow rate and
temperature that are constant at all points in the apparatus. These maintain a fixed value when
steady state is reached.

Flow

Time

Composition

xD

zA

xR

Time
2.3 Description of the operation of the distillation column

A distillation column is a gas-liquid contactor that allows the exchange of matter and heat
between the two liquid and vapor phases flowing in countercurrent. This device consists of a set
of trays as shown in Figure 2.2.

D, xD
−Q c

V1 Lo
11
V2 L1
2
V3 L2
3
Rectifying section

Vn Ln−1
n
Vn+1 Ln
F, zF Feed
Vm Lm−1
m
Vm+1 Lm

P’
Stripping section

V2′ L′3

V1′ L′2

Vo′ L′1
+Q B
B, xB

Figure 2 : Tray distillation column

 - The feed or mixture to be distilled is introduced at a certain level of the column. Two very
important pieces of equipment are used; one is located at the top of the column and which
corresponds to a condenser in order to condense the vapor flow leaving the first tray and
the other at the bottom of the column is the reboiler which allows to vaporize the liquid
leaving the last tray.
- The ratio (r = Lo/D) is called reflux ratio where Lo represents the flow of the liquid phase
that leaves the condenser and returns to the column at the level of the first tray. This flow
is obtained by removing the quantity of heat QC from the vapor phase (V1) that leaves the
first tray and enters the condenser.
- The ratio (θ = V0'/R) is called reboiling ratio where V0' represents the flow of the vapor
phase that leaves the reboiler and returns to the column at the level of the last tray (or first
tray from the bottom of the column). This flow is obtained by providing the quantity of
heat QB at the boiler.
- The section of the trays located above the feed constitutes the enrichment zone where the
vapor is rich in the most volatile constituent.
- The section of the trays located below the feed constitutes the depletion zone where the
liquid
is poor in the most volatile constituent.

3. Balances

3.1 Overall balances around the column

In steady state, which is the case of continuous operation of a distillation column, are
formulated as follows : "the quantities of total or partial material (relative to a constituent) and
heat entering and exiting in part or all of the column are equal".
a- Mass balance
With respect to the entire column, we can write :

F = D + B (1.1)

Partial balance with respect to the most volatile constituent, we find :

zF . F = x D . D + x B . B (1.2)
We find :

zF − x B
D= F. (1.3)
xD − xB

x D − zF
R= F. (1.4)
xD − xB

With:

F : Feed flow rate.

D : Distillate flow rate.

B : Bottom flow rate.

zF : Composition of the most volatile constituent in the feed.

xD : Composition of the most volatile constituent in the distillate.

xR : Composition of the most volatile constituent in the bottom.

b- Heat balance
F . HF + Q B = D . hD + B . hB + Q C (1.5)
With :

HF : Enthalpy of the feed.

QB : Quantity of heat to be supplied to the boiler.

QC : Quantity of heat to be extracted from the condenser.

hD : Enthalpy of the distillate.

hB : Enthalpy of the bottom.

3.2 Balances around the condenser

−Qc
hD , D, xD

H1 , V1 , y1 hD , Lo , xD
1

Figure 3 : Total condenser

a) Material balance

For a total condenser, we can write :

V1 = Lo + D (1.6)

y1 . V1 = xD . Lo + xD . D (1.7)

b) Heat balance

H1 . V1 = hD . D + hD . Lo + Q C (1.8)

After rearrangement, we find the quantity of heat to be extracted at the condenser per mole
of distillate :

QC Lo
= ( + 1) (H1 − hD ) = (r + 1)(H1 − hD ) (1.9)
D D

3.3 Balance around the boiler

1′
h1′ , L′1 , x1′ Ho′ , V1′ , yB
+QB

hB , B, xB

Figure 4 : Partial boiler

a) Material balance
L′1 = Vo′ + B (1.10)

x1′ . L′1 = yB . Vo′ + xB . B (1.11)

b) Heat balance

h1′ . L′1 + Q B = hB . B + H1′ . Vo′ (1.12)

The amount of heat to be supplied to the boiler per mole of bottom :

QB Vo′ L′1
= hB + H1′ . − h1′ . (1.13)
B B B
Ho′ : Enthalpy of vapor phase (Vo′ ) leaving the boiler.
h1′ : Enthalpy of liquid phase (L′1 ) which leaves the first plate from the bottom of the column.
We can use also the equation (2.5) :

QB D QC F
= . hD + hB + − . HF (1.14)
B B B B
3.3 Balance around the feed
On prend le cas où une certaine quantité de l’alimentation est vaporisée (figure 2.5).

𝐲𝐅 , 𝐕𝐅 , 𝐇𝐅′

𝐳𝐅 , 𝐅, 𝐇𝐅

𝐱 𝐅 , 𝐋𝐅 , 𝐡𝐅

Figure 5 : Liquid-vapor feed

a) Material balance
F = LF + VF (1.15)
zF . F = xF . LF + yF . VF (1.16)

b) Heat balance

HF . F = hF . LF + HF′ . VF (1.17)

Avec :
h F = x F . h F1 + (1 − x F ). h F2 (1.18)
HF′ = yF . HF′ 1 + (1 − yF ) . HF′ 2 (1.19)
Où :

xF : Composition of the most volatil component in the liquid phase of feed.

yF : Composition of the most volatil component in the vapor phase of feed.
LF : Flow rate of liquid phase of feed.
VF : Flow rate of vapor phase of feed.
hF : Enthalpy of liquid phase of feed.
HF′ : Enthalpy of vapor phase of feed.
HF′ 1 : Enthalpy of the component A1 in vapor phase of feed.
HF′ 2 : Enthalpy of the component A2 in vapor phase of feed.
hF1 : Enthalpy of the component A1 in liquid phase of feed.
hF2 : Enthalpy of the component A2 in liquid phase of feed.

4 Determination of the stage number by Mc Cabe-Thiele method

4.1 Principe

The Mc Cabe-Thiele method is based on Lewis' simplifying assumptions. It is a graphical

method that allows to determine the number of theoretical stages in a distillation column. It was
first published by Warren L. McCabe and Ernest Thiele in 1925.This method is the most used
but not very rigorous, it can be applied to ideal or close to ideal mixtures.
The determination of the number of theoretical stages consists in the construction of a staircase
between the operating lines and the equilibrium curve where the points of the equilibrium curve
correspond to the streams leaving the stages in equilibrium and the points of the operating lines
to the streams that cross between the stages.

4.2 Lewis Hypotheses

Lewis' simplifying assumptions are as follows :

1- The column is perfectly insulated (adiabatic).

2- The heats of mixing of the two components of the binary are negligible (zero heat of mixing).

3- The molar heats of vaporization of the two components are equal.

4- The liquid-vapor equilibria are achieved on each plate (theoretical stage).

Remarque :

Both assumptions 2 and 3 imply that the flow rates of liquid and vapor do not change throughout
the column (i.e., constant molar overflow).

V1 = V2 = ⋯ = Vn = V
{ Rectifying Section }
Lo = L1 = ⋯ = Ln = L
Et :

Vo′ = V1′ = ⋯ = Vm
′
= V′
{ Stripping Section }
′ ′ ′ ′
L1 = L2 = ⋯ = Lm = L
4.3 Operating lines
a- The rectifying section
If we do a mass balance between a plate P and the condenser, we find :

0 D, xD
1

P
VP+1 , yP+1 LP , x P

n
Feed

Figure 6 : Rectifying section.

VP+1 = LP + D (1.20)

yP+1 . VP+1 = xP . LP + xD . D (1.21)

So, we find the equation of the rectifying section operating line :

L L
yP+1 = . xP + . xD (1.22)
V D

xP : Composition of the liquid leaving the P plate of the rectifying section.

yP+1 : Composition of the vapor leaving the P+1 plate of the rectifying section.

Equation (1.22) can be written as a function of the reflux ratio r by :

r xD
yP+1 = . xP + (1.23)
r+1 r+1
Slope Origin

With :
L
zF ≤ x P ≤ x D , r=
D
Remark :

By replacing xP with xD we find (yP+1 = xD ), the rectifying section operating line pivots
around the point D(xD, xD) which is fixed and its direction varies according to the reflux ratio r.
 The rectifying section
operating line D

0 x
D 1
x
Figure 7 : The rectifying section operating line.

b- The stripping section

The material balance between the plate P and the boiler, we find :

Feed
m

vP′ , yP′ L′P+1 , xP+1

′

1
B, xB 0

Figure 8 : Stripping Section.

L′ P+1 = V ′ P + B (1.24)
′
xP+1 . L′ P+1 = yP′ . V ′ P + xB . B (1.25)

So, we find the equation of the stripping section operating line :

L′ ′ L′
yP′ = . x − .x (1.26)
V ′ P+1 B B

′
xP+1 : Composition of the liquid leaving the plate P+1 of the stripping section.
yP′ : Composition of the liquid leaving the plate P of the stripping section.

Equation (2.26) can be written as a function of the reboiling ratio θ by :

θ+1 ′ xR
yP′ = . xP+1 − (1.27)
θ θ
Avec :
V′
xB ≤ xP′ ≤ zF , θ=
B
Remark :
′
Replacing xP+1 by xR we find (yP′ = xR ), the stripping section operating line pivots around
a fixed point B(xB , xB ).

The stripping section

operating line

0 xB 1
x

Figure 9 : The stripping section operating line.

c- The feed line

For a given reflux ratio at the head of the column, the two operating lines (rectifying section
and stripping section) meet at a point I located between the equilibrium curve (y = f(x)) and the
first bisector (y = x), the location of point I varies according to the reflux ratio r. Thus the
resolution of the two equations (1.28) and (1.29) gives us the point of intersection of the two
operating lines and which verifies the equation of a third line called the feed line.
 L L
y = . x + . xD (1.28)
V D
L′ L′
y = ′ . x − . xB (1.29)
V R

We multiply equation (1.28) by V, and equation (1.29) by V', and we subtract the two
relations we find :

y. (V − V ′ ) = x. (L − L′ ) + (xB . B + xD . D)

We know that :
zF . F = x D . D + x B . B
⇒ y. (V − V ′ ) = x. (L − L′ ) + zF . F

(L − L′ ) F
y= .x+ .z (1.30)
′
(V − V ) (V − V ′ ) F

Knowing that for a liquid-vapor feed (F = LF + VF ) :

L′ = L + LF (1.31)
Et :
V = V ′ + VF (1.32)

Substituting equations (1.31) and (1.32) into (1.30) we find :

−LF F
y= .x+ .z (1.33)
VF VF F

On pose :
VF
β= (1.34)
F

β : Feed vaporization ratio.

0≤β≤1

Equation (1.35) represents the feed line as a function of the feed vaporization ratio β:

β−1 zF
y= .x + (1.35)
β β
Note :
By replacing by zF we find that : y=zF, the feed line pivots around a fixed point F (zF,zF)
depending on the state of the power supply.

Feed line

0 zF 1
x

Figure 10 : Feed line.

The equation of the feed line can be written as a function of a coefficient q which represents the
fraction of the feed which descends in liquid form :

L′ − L V − V′
q= ⟹ 1−q=
F F
Equation (1.30) of the feed line becomes :
q zF
y= .x − (1.36)
q−1 q−1

With :
β=1−q (1.37)

According to the heat balance relative to the feed stage (figure 1.11) :

HF . F + hL . L + HV′ . V ′ = hL′ . L′ + HV . V (1.38)

Applying Lewis's assumptions :

h L = h L′ et HV = HV′

HF . F = hL (L′ − L) + HV (V − V ′ )
 ⟹ HF . F = q. hL + (1 − q). HV

V, HV L, HL

F, HF

V ′ , HV′ L′ , HL′

Figure 11 : Feed section.

We finf :
HV − HF
q= (2.39)
HV − hL

where q is the thermal factor of the feed.

4.4 Feed states and direction of the feed line

The feed can be in several forms (liquid, vapor, or liquid-vapor) depending on its temperature.
With :
TF : the temperature of the feed.
TB : the boiling temperature of the feed.
TR : the dew point temperature of the feed.

a- Subcooled liquid feed (TF < TB)

The feed is a subcooled liquid that enters

the column at the feed tray with a temperature
L V
lower than its boiling temperature, it descends
into the stripping section and causes the condensation
of a certain quantity of the vapor present on the tray F
so that the temperature of the feed reaches that of the
feed tray :
L′ = L + A + L′′
L′ V′
′ ′′
V=V −L
L′′ : the quantity of condensed vapor.
b-Saturated liquid feed (TF=TB)
L V
If the feed is in the form of a saturated liquid,
all the introduced flow will go back down into
the stripping section. In this case, we find : A
L′ = L + F
V′
V=V ′ L′

c-liquid-vapor feed (TB < TF < TR)

In tha case of a liquid-vapor feed, the mixture L V

will be distributed in each of the sections, therefore :

F
′
L = L + LF
V = V ′ + VF
L′ V′

d- Saturated vapor feed (TF=TR)

In the case of saturated stream, the feed is completely L V
steam at the dew point temperature TR, it rises in the rectifying
section :
A
L′ = L
V = V ′ + VF
L′ V′
e- Superheated vapor feed (TF>TR)
The feed enters the column with a temperature
L V
higher than the TR and which will cause the vaporization
of part of the liquid so that the TF reaches that of the column :
A

L′ = L − V ′′
V = V ′ + F + V ′′ L′ V′
V ′′ : the amount of liquid vaporized.
 1

a
y b
c
d
F
e

0 zF 1
x

Figure 12 : Representation of the different states of the feed depending on its temperature.

Feed state TF<TB TF=TB TB<TF< TR TF=TR TF>TR

q >1 1 ]0,1[ 0 <0

β <0 0 ]0,1[ 1 >1

Slope of the feed line >0 ∞ <0 0 >0

Table 1.1 : Representation of the different states of the feed depending on its temperature.

4.5 Graphical construction of theoretical stages

To construct the theoretical stages, we can follow the following steps :

1- Place the points : D (xD,xD), R (xR,xR), and A (zF,zF).
2- Set a reflux ratio r at the top of the column.
3- Plot the rectifying section operating line.
4- Plot the feed line.
5- Plot the stripping section operating line by joining points R and I.
6- Build the theoretical stages in a staircase between the equilibrium curve and the two
operating lines starting from point D until point R is exceeded.
 D
y= f (x)

Boiler

R
0 xR xI zF xD 1
x

Figure 13 : Construction of theoretical plates using the Mc Cabe -Thiele method.

According to figure (1.13), there are 10 theoretical plates, distributed as follows :

05 rectifying section plates.

01 feed plate.

03 stripping section plates.

01 boiler.

4.6 Minimum number of theoretical stages Nm

The minimum number of stages can be determined graphically using the equilibrium curve,
or with an analytical method (Fenske relation).

4.6.1 Graphical method

The column operates at total reflux where r→∞ therefore (D=0), the two operating lines
coincide on the first bisector with a slope equal to 1. The minimum number of plates N m can be
determined by constructing the stages in a staircase between the first bisector and the equilibrium
curve (figure 1.14).
 1

1 D
2

y 3

5
B
0 xB xD 1
x

Figure 14 : Construction of the minimum number of stages.

4.6.2 Analytical method (Fenske equation)

In the case where the reflux ratio r tends to infinity, the operating lines of the rectifying and
stripping sections coincide on the first bisector. Therefore, we can write :

yP+1 = xP

The liquid-vapor equilibrium at a P+1 plate can be represented by the equation:

yP+1 = K P+1 . xP+1 (1.40)

Plate 1 : y1 = x0 = K1 . x1 (Total condenser ⟹ y1 = xD)

Plate 2 : y2 = x1 = K 2 . x2
Plate 3 : y3 = x2 = K 3 . x3
. .
. .
. .
. .
Plate P+1: yp+1 = xp = K p+1 . xp+1
. .
. .
. .
Plate n+1 : yn+1 = xn = K n+1 . xB
After rearrangement we can write :
xD = K1 . K 2 … … … … . K n+1 . xB

(1 − xD ) = K1′ . K ′2 … … … … . K ′𝑛+1 . (1 − xB )

Therefore :
xD K1 . K 2 … … … … . K n+1 xB
= ′ ′ ′ .
(1 − xD ) K1 . K 2 … … … … . K n+1 (1 − xB )
With :

xD xB Kn
= α1 . α2 … … … … αn+1 . , αn =
(1 − xD ) (1 − x B ) K ′n

αn : Relative volatility of the most volatile component compared to the least volatile component
in
tray n.
K n : Equilibrium constant of the most volatile component in tray n.
K ′n : Equilibrium constant of the least volatile component in tray n.

For an ideal mixture, we can consider a constant value of the relative volatility equal to αmoy .
After rearranging, we find equation (1.41) which represents the Fenske relation to determine the
minimum number of theoretical plates :

x 1−x
ln {(1 −Dx ) ( x B )}
D B
Nm + 1 = (1.41)
ln αm

4.6.3 Application example

An ideal binary mixture (propanol-1+isopropanol) with a molar fraction of 0.35 in propanol-

1 must be rectified in a plate column equipped with a total condenser and a partial reboiler in a
distillation column operating continuously to obtain a distillate with 93% mol of A and 3% in the
bottom. The average relative volatility is equal to 2.18.

0,93 1 − 0,03
ln {(1 − 0,93) ( 0,03 )}
Nm + 1 = = 7,779
ln 2,18

Nm + 1= 8 theoretical stages (07 plates + the boiler)

4.7 Minimum reflux ratio

The minimum reflux corresponds to the case of a column with an infinite number of plates.
Graphically, this situation corresponded to the intersection or tangency of an operating line with
the equilibrium curve. Therefore, the meeting point of the operating lines and the feed line is
located on the equilibrium curve where the rectifying of the vapor phase as well as the stripping
of the liquid phase are zero, the separation of the constituents requires an infinite number of
plates, in this case the value of the reflux ratio is minimal and equal to r m.
Determining the value of the minimum reflux ratio is very important since the value of the reflux
rate imposed at the top of the column must be greater than this value for the column to operate.
The value of the minimum reflux ratio rm can be calculated in two ways : from the value of the
ordinate at the origin and the slope in the equation of the The rectifying section operating line
(eq 1.22) :
a- The origin :
When (x=0), we can read graphically the value of yom (figure 1.15) :
xD
y om =
rm + 1

So :
xD
rm = −1 (1.42)
yom
b- The slope:
In the case of a liquid-vapor feed of composition z F, the value of rm can be calculated using the
slope of the line (DI) :

rm xD − yI
= (1.43)
rm + 1 x D − x I

I(xI , yI ) is the meeting point located on the equilibrium curve (figure 1.15).
 1

D
y= f (x)

y
I
yI

yom F

0 xI zF xD 1
x
Figure 15 : Minimum reflux ratio rm