UNIT OPERATIONS COURSE ADSORPTION

ABSORPTION

The absorption of dissolved gas in a liquid solvent containing one or

several reagents are found in four fields of industrial application:

Washing of acid gases: CO2, H2S, CO, SO2, NO,... for the manufacturing of
synthetic gas and to combat air pollution etc...
Fabrication of pure products: H2SO4, HNO3BaCl2,…
Biological processes: fermentation, sludge oxidation, production of
proteins from hydrocarbons, biological oxidations etc...
Liquid phase processes: Oxidations, hydrogenations, sulfonations,
nitrations, etc...

Reminders

1.1 Solubility of gases

Let's consider the case of a binary gas-liquid system:

If a gas and a non-volatile liquid are in equilibrium, then the concentration

the gas dissolved in the liquid is called the solubility of the gas in that liquid
at the existing temperatures and pressure then:
The main characteristic of gas solubility is that it is
necessary for maintaining a given concentration of gas in the
liquid phase, that this gas exerts in the gaseous phase in contact,
a partial pressure whose absolute value is well defined. Conversely,
one can say that the dissolved gas exerts a partial pressure above
the solution.
The solubility of gases decreases rapidly with temperature.

1.1.1 Ideal liquid solutions

The characteristics of an ideal solution are:

a- The forces of attraction and repulsion intramolecular in the solution

are exchanged when mixing the components.
The volume of the solution varies linearly with composition.

1
UNIT OPERATIONS COURSE ADSORPTION

There is neither adsorption nor evolution of mixing heat of the

constituents.
d- The vapor pressure of the solution varies linearly with the
composition expressed in mole fraction.

In this case, we can use Raoult's law:

p* p vx

with p* partial pressure of the gas in equilibrium with the liquid

pv vapor pressure of the solute gas
x mole fraction of the gas in the solution

1.1.2 Henry's Law - Non-ideal solution

For non-ideal dilute solutions, one can consider that the fraction
molarity of solute y* in the gas in equilibrium with a solution whose
mole fraction in solute is x, is written as:

p*
y* mx
pt

1.2 Coefficient of volumetric transfer conductance

Let's consider the absorption in which a gas is dissolved in the liquid without
that there is a reaction. The basic representation of the transport of solute gas is
based on the concept of the additivity of resistances in the gas phase and in the phase
liquid, it is also assumed that the interfacial resistance is negligible.
The overall absorption flux per unit volume of absorber is then:

N A.a k G.a  p G p L  
k L.a C *  C L 
With
a interfacial area between the gas and the liquid referred to the unit of
volume
Na average specific absorption flux per unit area
interfacial
Pg, pI partial pressures of the soluble gas respectively within the
Look at the interface
C* and CL dissolved gas concentrations at equilibrium respectively at
the interface and within the liquid.
kg mass transfer coefficient in the gas phase. One can
attach it to a resistance of a stagnant gas film

2
UNIT OPERATIONS COURSE ADSORPTION

of thickness gthrough which the gas is transported

exclusively by molecular diffusion, the composition at
his gas being uniform.

For the transfer of matter in the liquid phase, several models based on
distinct hypotheses regarding the behavior of the liquid lead to some
different expressions of the mass transfer coefficient kL.

N A.a K G.a p  p*
G  K L .a C*L C L  K G.a.P y y *  K L.a.C x *  x 
KGand KL global mass transfer coefficients

The knowledge of volumetric coefficients is necessary for calculation

of the sizing of the devices. Unfortunately, due to the complexity
variations of 'a' and subsequently of that of the volumetric coefficient, it has been
impossible until now to obtain a theoretical expression for the
coefficients. It is therefore necessary to resort to empirical expressions.

2. Continuous contact of immiscible phases

The absorbers used come in very diverse shapes:

tubular reactors with bubbles, droplets, falling film, packing

plateaus, etc...
mechanically agitated tank reactors,
jet or venturi type engines.

These different geometric shapes are due to competition between

phenomena :

of chemical thermodynamics (solubility of gases, diffusivity of

reactants);
transfer of matter and heat on each side of the gas interface -
liquid (transfer coefficients and interfacial areas);
from chemical kinetics (rate constants, order of reaction selectivity,
etc…)

2.1 Equipment used during gas-liquid operations

3
UNIT OPERATIONS COURSE ADSORPTION

a- Progressive devices: Filling columns

Gas
Liquid

Gas

Liquid

The contact is continuous between the liquid and the gas. The pressure losses are
less significant and fluid retention is lower.

b- Stacked devices: plate columns

Liquid
Gas

Gas

Liquid

The contact is intermittent, layered.

c- Empty columns

Liquid
Gaz

Gas

Liquid

2.2 Equipment used during liquid-liquid operations

4
UNIT OPERATIONS COURSE ADSORPTION

a) Mixers with separators

Liquid A

Liquid B

Liquid A

Mixture Liquid B

Separator
b) Packing column with a blower

Heavy B liquid
Piston
This equipment
Allows to accelerate
Pulse the transfer of matter

Light Liquid A

Liquid B

In general, A and B are not completely immiscible.

3. Filling columns

Packed columns are used in the industry during operations.

absorption, distillation, and extraction.

These columns create a continuous contact in co-current or counter-current of both.

gas and liquid phases. The exchange between the two phases occurs all the better
that we have a large contact surface and good turbulence.

5
UNIT OPERATIONS COURSE ADSORPTION

The efficiency of packed columns depends on the parameters.

following:

The characteristics of the upholstery.

The different types of flow.
The hydrodynamic conditions.
Retention

3.1 Choice of padding

A padding must be such that:

The specific surface (surface volume). Must be the largest.

possible.
It must have a large volume of voids with a high porosity in order to
to have the lowest possible pressure drop (the resistance to
the lowest possible flow.
It allows for a uniform distribution of the liquid.
It must be of low density, for the large cladded columns, the
the weight of the filling can cause serious problems of
support.
It must be resistant to corrosion and relatively inexpensive.

The padding plays an important role because:

It increases the liquid-gas contact area.

It creates micro-whirlwinds in the liquid film, which
allows for the renewal of the surface and promotes mixing and
the local homogenization of the liquid concentration.
It prevents the formation of large convection currents.
longitudinal in the gas phase, which maintains a difference
of concentration between the inlet and outlet of the column.

3.2 Different types of filling

6
UNIT OPERATIONS COURSE ADSORPTION

Among the rings, we can mention:

D
Raschig (1830)

Ceramics, porcelain
Steel, plastic H
Height = Diameter
Lessing: ceramics, porcelain

H=D D < 50 mm: bulk

D > 50 mm: stacking

Pall: Ceramic, steel,

important porosity, good

contact, increase of the
turbulence. It includes some
tabs folded verse
the inside (4 Tabs)

The Selles
Berlin Ode
oIntalox

Spherical balls:

Good contact, low porosity and

large pressure drop.
A support is needed to hold the
padding

Ag 6
ag
Vgd g

The choice of filling depends:

7
UNITARY OPERATIONS COURSE ADSORPTION

The system to be studied.

About the cost.
Of efficiency.

Remark
Yes we can accept a higher gas flow
3.3 Types of flow

3.3.1 Watered Flow

The gas phase is continuous while the liquid phase is discontinuous.

this is for or against the current. The support grid for the padding is placed after the
liquid distributor.

The liquid drips over the padding

Liquid there is a liquid film, the diameter
the pores decrease hence the surface
Assigned to the gas flow
Droplets decreases what will lead to a
greater gas speed.
Gaz

If we keep the liquid flow rate L constant and vary the flow rate of
gas G, by increasing gas G will push back the liquid, the liquid will occupy everything
the volume of the void and the gas can no longer pass, and we observe the phenomenon of
the congestion.

3.3.2 Drowned flow

In this case: the liquid phase is continuous, the gaseous phase being
discontinue to either co or counter-current. The liquid occupies the entire section offered by
the filling and gas disperse in the form of bubbles.

If the flow rate G increases, the size of the

Liquid sparkling bubbles will increase and create
a path by pushing back the liquid that
Bubbles will no longer be able to pass and we observe
the phenomenon of congestion.

Gas

8
UNIT OPERATIONS COURSE ADSORPTION

Note: In the cases of co-current flow whether downward or

Upward, there is no congestion.

3.3.3 Downward co-current flow

If one of the flows is increased, we reach the state of the irrigated flow,
drowned or pulsed characterized by an alternating passage of gas and liquid, or
the flow in fog because the liquid will form a very thin layer on the
filling then very fine droplets.

Liquid Bubbles L
Of gas G
L

Rosé Drowned Pulsed Fog

Optimal transfer

4. Hydrodynamics of solid-liquid-gas systems

4.1. Pressure losses of the gas flow

The pressure drop in a packed column is influenced by both

gas and liquid flow rates.

Ln P (1) (2) (3)

L G
(III)
B

(II)
A

(I)
LnG

The flow of gas through a packed column is generally turbulent.

hence the slope of the curve (3) is equal to (II). (Prediction of the loss equation
of load in turbulent regime).

For a constant gas flow G, the pressure drop increases with

the increase in liquid flow.

9
UNIT OPERATIONS COURSE ADSORPTION

Let's consider curve (I), with a constant liquid flow L:

Up to point A, the pressure drop is similar to that of the filling.
sec (3).
The loss is the same; only the pressure drop is high.
At point A, a change in slope occurs indicating a
faster increase of pressure gradient with speed of the
gas.

Point A is called the charge point.

Description of the different areas

Zone I: The liquid-gas current is weak, hence we have a low

efficiency: the liquid-gas interaction is negligible.

For a water-gas system, and for regimes lower than the point of
charge A Leva established the following correlation:

P G2
.10 L .
Z g

L and G: mass flow rates of the liquid and gas (kg/m)2. s).
, Tabulated constants.
P
Z
pressure drop per unit height of the packed column
(mm Hg/m column)

This equation is valid when:

1- the diameter of the particles is at most equal to 1/8 of that of the

column
2- the stacks are unordered; the degree of compaction can have a
great influence;
3- the liquid flow rates lower than those corresponding to the point of
the congestion.

Zone II: Loading area

10
UNIT OPERATIONS COURSE ADSORPTION

At point A, there is a change of pace, the friction or braking is

then sufficient to hinder the descent of the liquid causing a retention of the
liquid, what decreases the free space available to the gas, is the charge point,
(difficult to observe experimentally).

In this area, the Ergun equation applied to the flow is used.

single-phase.

P
h .G h .G2
Z
P
Knowing Z
in the case of a packed column (dry packing case), we
can deduce the pressure drop for a liquid L flowing:

 P L,G f L P 0,G

fLbeing a corrective factor based on the flow L, there are tables providing:

fL f  L  , stuffing

Zone III: Engorgement

If we continue to increase G, the fluid retention increases quickly.

and we reach the second characteristic point B, where the pressure drop increases
suddenly to dizzying proportions. The speed of the gas is
large enough at that moment to prevent the liquid from flowing out
the effect of gravity, and force it to accumulate in more or less layers
thickened in various places of the column. It is the point of congestion and the
column can no longer function. It is important to note that the point
congestion is not for a flow or speed of liquid (or gas)
but rather for an association or combination of two flows G and L
producing a vertical increase in pressure. There is a good
graphical correlation of congestion points:

L G
Log B
On the x-axis:A .G
L

ap 1 L
On the vertical axis: B G2. 0.2
L
. .
G. Lg c
3

engorgement

11

Log A
UNIT OPERATIONS COURSE ADSORPTION

Where ap: specific surface

porosity
L: viscosity of the liquid

4.2 Retention

The retention is the volume of liquid held by m3of decorated column.

This value is important in the case of absorption with chemical retention.
and in distillation.

We observe two types of retention: static and dynamic.

a- static retention

It is the volume of liquid that is trapped in the lining at flow rate of

zero gas, after the establishment of the regime.

b- Dynamic or operational retention

It is the volume of liquid held by a certain gas flow rate. Thanks to some
volume measurements, we can draw two diagrams.

m 3water
L
m 3garnished column

L L3L2L1 L G3G2G1

G(kg/m3.h) G(kg/m3.h)

L3L2L1 G , L

Correlations
0.6
L
water 0.0011
dp

12
UNIT OPERATIONS COURSE ADSORPTION

6 1  
with : dp 3
.N

L: water flow rate (kg/m2.h)

N : number of units or filling elements per m3of decorated column.
diameter (in mm) of the sphere equivalent to a packing element.

 o n e   .V N.V 1

V: Total padding = 1m3

V1Filling

dp3 6 1  
 1  N. dp3
6 .N

If the liquid is not water, we use the following correlation:

0.1 0.73
liquid liquid water water
. .
water water liquid liquid

with : surface tension

water73 dynes/cm

n : constant function of liquid flow

n

L (kg/m3.h)

5. Material balance

Let there be an absorption column represented by the following diagram:

L2, x2
L and G are flow rates of liquid and gas.

G, y G2" and y"2 Index 1: bottom of column

Index 2: top of the column

13
UNIT OPERATIONS COURSE ADSORPTION

L, x y: fraction in the gas phase

x: fraction in the liquid phase
G1, y1
In the case of G absorption2< G1
L1, x1 L2< L1

The total balance gives:

G1+ L2= G2+L1

L1L2= G1– G2

Assessment of the active ingredient:

G1. y1+ L2.x2= G2.y2+L1.x1

At any level:

G + L2= G2+L
G.y + L2.x2= G2.y2+ L.x

L G 22-L2 .x2
y x
G G

G L 2 G G 2 y 2 L 2 x 2
y 2
x
G G

L
Note: This is not a straight line because is variable and function of height.
G

This equation is the locus of points whose coordinates are the

concentrations of the phases that intersect at any level in the
column.
The graphical presentation is called operation line or outline.
operational.

In the case where L and G are inert, thus constant flows along
the column (y and x, being molar ratios) then the operating line is a
right.

1 – G, Y1, X2, and X1known

The construction of a bridge X1two unknowns, Y2and L.

14
UNIT OPERATIONS COURSE ADSORPTION

a- SiY1= Y2
L
0 L=0
G

The line AB is horizontal.

b- SiY2= Y2*

There is a gas-liquid equilibrium at the top of the column. The slope of the line
L Y1 Y2*
operative G X1 X 2
this implies that the column must be infinite
max

high in general L = a.Lmax, with 0 < a < 1

Y L
0
B G A
Y1
L
G max

Y* X
X2 X1

2 –G, Y1, X2, and Y2unknown

Purification of a gas, L and X1are the unknowns.

Y1 B E
L Y1= mX1* balance between
The two phases
G G min

Y2 A Y*=mX

X
X2 X1*

15
UNIT OPERATIONS COURSE ADSORPTION

a-/ SiX1= X2 line BA

L
L= the operative line is vertical
G

b-/ SiX1= X1*

L Y1 Y2
The liquid is in equilibrium with the vapor G L min
min X1*  X 2
In general L = a.Lminwith a >1.

Lmin, we have the saturated liquid. If a is too large, the transfer is

The case where L
bad, the solution is too diluted.

Note: the case where the equilibrium curve is convex.

The flow of the liquid L

the gas flow G
is determined graphically in such a way that the
min
min

operating line is tangent to the equilibrium curve.

Y
L
Y1 Operational slope law G min

Equilibrium curve

Y2
X
X1 X2
3 Recycling problem
What is the amount of recycled liquid?

L, x2 L' = L" +L

L', x2'
Inside the envelope
G, y2 in dotted lines
L.x2+ G.y1= G.y2+ L.x1

L" Inside the column

L'.x2' + G.y1= L.x
'
1+ G.y2
G, y1

16
UNIT OPERATIONS COURSE ADSORPTION

L, x1
liquid mixer at the inlet

L + L" = L'
L.x2+ L".x1= L.x '
2
'

L.x2+(L' 1= L.x
'
2
'

L' L'
x 2  1 .x1 .x2 '

L L
L'
recycling rate
L
L' x2 x1x1 x 2
L x'2 x1x1 x '2
y

y1 The maximum recycling rate

L' x1 x 2
L max
x1 x *2
y2
L' L'
b with b > 1
L L max

x2 x2* x1 x

Domain of variation of x2'

The maximum recycling rate:

L' x1 x 2
L max
x1 x *2
' '
L L
b with b <1
L L max

4- Sizing, calculation of the height of an absorber

The calculation of the height of an absorber depends on the type of contact:

Stage absorbers (plate columns)

Continuous absorbers (packed column).

17
UNIT OPERATIONS COURSE ADSORPTION

The efficiency of the column is not calculated in the same way because
that the contact is different.
4.1- Stage contact absorbers

The height is equal to the product of the number of actual plates by

the spacing between two plates (e)

H N Pr.e

The number of real plates is the quotient of the number of plates.

theoretical on the overall efficiency of the column

N
N Pr
E
The number of theoretical trays is determined by the graphical method.
from Mac Cabe and Thiele based on the plots of the curves.

Equilibrium curve
Operational right
The triangles representing the plateaus.

XN-1

YN
The points (YN, YN+1); …; (Y0, X1) are the
N from the operational right.
XNYN-1
Xi+1 YI (Y1, X12, X2); (YN, XN) belong
to equilibrium curve.
XiYi-1
X2Y1
1
Y0

Y X1

Y0

1
Y1
We number the floors from top to bottom
(by convention)
2

Y2
X 18
YN
XN+1 X3 X2 X1
UNIT OPERATIONS COURSE ADSORPTION

We end up with a sequence of stairs.

between the two operational lines and
of balance.

The number of steps gives the number

theoretical plateaus.

For the liquid to be in equilibrium with the gas, it is necessary that the time of
make it large and the contact surface should be large too this is not
possible in practice.
G, Yi-1 G, Y1
Plateau i
L, Xi L, Xi+1

If the platform is theoretical then Xiand Yiare in balance.

Calculation of the plateau fraction

Y YN1 YN
fraction
YN1 YN'

The number of trays

N-1 theoretical will be fractional
YN-1
but the number of trays
YN reals is an integer.
YN'
N Pt
X N Pr and we round off.
E
XN+1 XN XN-1

4.2 - Continuous contact absorbers (with filling)

Let it be a differential element of height dh. The transfers of matter in

these volumes are described by the following balances:

19
UNIT OPERATIONS COURSE ADSORPTION

Gy

dm d  Gy  d Lx   N A a  dV

dh
dm When ydG Ldx xdL

Lx
For low transfers: dL = 0
dG = 0
y
yi When Ldx
xI
x d Gy  ky y y i  dA
d Lxy kx x i x  dA

with dA a.Sdh

It is assumed that the flow (NA.a) is constant at every point of the element
differential.

If there is a transfer of a single substance, we will also have:

dG dL  N A.a  S.dh

The flow NAIt is expressed in moles / unit of time and volume.

dG d  Gy  ydG When
dy
dG G
1 y
G.dy
 N A.a  S.dh
1 y
L.dx
likewise:  N A.a  S.dh
1 x

a being the interfacial surface area for mass transfer per unit volume filled
(m2/m3) a padding.

a = f(a dry padding, L, G and L).

a- Cases where the gaseous film controls the transfer rate

The transfer rate is given by:

20
UNIT OPERATIONS COURSE ADSORPTION

 N A.a  k G.a P P i
 k G.a.P y y i 
K G.a.P y y * 
G.dy
K G.a.P y y *  .S.dh
1 y
y G 1 dy
h
y12 SKG.aP1 y   
y y * 
1y 1 d Gy 
y2 S KG.a.P y y
1 *


b- cases where the liquid film controls the transfer rate

In this case, we use one of the equations containing the potential expressed in
liquid phase concentration units.

 N A.a  k L.a C i C  K L.a C *  C 

k L.a x I x  K L.a x *  x 
L.dx
1  dx
 K L .a.C  x *  x
 .S.dh
x L 1 dx
h 
x1 2 SK .a.C1 x
L

  x*  x 
x 1 d  Lx 

x1 2
K L.a.C.S  1  x  x*  x  
C is the total concentration in moles per unit volume.

c- General case where both resistors are important

G L
h
y S dy S x dx
y 2 K .a.P  1  y   y y *
1
G
 x 2 K .a.C  1 x   x *  x
L
1

These equations are valid both in the case of counter-diffusion.
equimolar only in the case of diffusion through a stationary phase (not
diffuser).

Let's examine these equations in more detail:

Equimolar back diffusion case

( This case occurs during the rectification distillation where there is an equal rate of
transfer of the more or less volatile constituent between the phases.

In this case:

21
UNIT OPERATIONS COURSE ADSORPTION

h dy
y dy
y
h dh G G
0 y1 k G.a.S.P.  y i y
2
 y1 K G.a.S.P. y y *
2
 
The molar flow rate G is considered constant:

For the same assumptions,

h dx
x x dx
h dh L L
0 x1 k L.a.S.C.  x  x I
2
 x1 K L.a.S.C. x x *
2
 
In the case of diluted gases, the absorbed component is present in
the gas in low concentration, even at the entrance of the column and that the solutions
the components that are found in the column are sufficiently diluted,
terms in (1-y) and (1-x) can be considered equal to one.

Other coefficients K can be used.y.a and Kx.a, instead of kG.a and kL.a.

G G G
yS dy S y dy S y dy
h 1
y
 *
y 2 K .a y y  K y.a y2  y y *
1
 K y.a y2  y y I
1

HG NG H OG N OG

L L L L
x S dx xS dx S dx
x S xdx
h
x 2 k1 .ax x
x i

 *
x 2 K .ax  x
1
x

 k x.a x 2 1 x  x
i  K x.a x2  x  x
1 *

HL NL H OL N OL

H = height of transfer unit [m]

N = number of transfer units [-]

G L
HG S HL
k y.a k x.a

G L
H OG S H OL S
K y.a K x.a

y1 dy x1 dx
NG NL
y2 y y I x2 xx
i

22
UNIT OPERATIONS COURSE ADSORPTION

y1 dy x1 dx
N OG NL
y2 y y*
x2 x x
*

k(x, y) Partial mass transfer coefficient (liquid side, gas side)

K(x, y) Global mass transfer coefficient
(ky.a) Mass transfer coefficient kmole/m3.s
(y-yi) and (xi-x) Difference in partial exchange potential on the gas side and the other side
liquid
(y-y ) and (x -x)
* * Difference in potential for global exchanges on the gas side and on the side
liquid

If the equilibrium curve is a straight line

y* = mx
y* = mx . b

The global coefficients:

1
Ky
1 m

k yk x Kx
Ky
1 m
Kx
1 1

k xm ky

These denominators are the sum of the resistances to transfer in phase.

gaseous and liquid.

Case of diffusion through a stationary component

The material transfer operations, where this case occurs, are the absorption of
gas, liquid-liquid extraction, ion exchange, and absorption.

The molar flow rate of inert gasV '  1  y  V

y dy dy
d  Vy  V 'd V' V
1 y  1 y  2 
1 y 
dy
d  Vy  V ky a y i y S dh
1 y
Ky 
ay *  y S dh
h y2 G dy
dh * 
a  
1 yy y

0 y1 S Ky

23
UNIT OPERATIONS COURSE ADSORPTION

G
y S dy
y1 k2 y a1 y 
  y i y 

In terms of liquid phase:

Lx L
h S dx x
S dx
dh
0 x1 k x a1 x2 
  x i x  x1 Kx 2
a  1 x   x *  x 
y2 y2
HG dNG H OG dN OG
y1 y1

x2 x2
HL dN L H OL dN OL
x1 x1

 1 y  Lm
On can multiplier and divide by yi
the equation
G
y S dy
y1 k2 y a1 y 
  y i y 

G  1 y  Lm
S dy k y a 1  y  Lm
where k 'y a who is independent of the
ky a 1 y  Lm 
y i y 1


concentration.

h
h
dh
G 2 y 1  y  Lm dy
0 ky a S 1  y  Lm mine
y1  1  y   y i y 

h
0
h
dh
Ky a
G
S 1  y  Lm m my
y1
2

1 yLmy  dyy 
y

 1  y   *

similarly for the liquid phase:

h
h
dh
L 2  1  x  Lm
x dx
0 kx  1 x
a S  Lm my
x1  1  x   x i x 

h
h
dh
L 2 x 1  x  Lm dx
0 Kx  1 x
a S  Lm my
x1  1  x   x*  x 

The Lm index is the logarithmic mean:

24
UNIT OPERATIONS COURSE ADSORPTION

x2  1 x 2    1 x 1  x 1 x 2
 1  x  Lm x1 1 x2 1 x2
Ln Ln
1  x1 1 x 1

y2 y1 y 2
 1 y  Lm
similarly for y1 1 y 2
Ln
1 y1

7- The concept of the transfer unit

The equations used for calculating the height of a packed column

can be rearranged to allow the introduction of a concept that throws a
new light on the functioning of the matter exchanger. Colburn and
y dy
Chilton (1935) proposed to call the integral y2 
 1  y  y y *
1
 the number
of transfer unit.

7.1 Determination of the number of transfer units (NUT)

NUT represents the separating power of an absorber. It is determined by

using the operational diagram.

y
Equilibrium curve

y1 Operational curve

Pent
yi

Y2

x2 x xix1 x

G dy L dx k y  y y i  k x x i x 

k xy y i

k yx x i

25
UNIT OPERATIONS COURSE ADSORPTION

For a column point (y, x) we can determine the concentrations at

x k
the interface (yI, xithanks to the intersection of the slope line  k passing through
y

(y,x) and the equilibrium curve y*=f(x).

In a gas absorption problem, the number of transfer units

east

N OG
y2

y1
 1 y dy
 1  yLm y  y *
NG
y2

y1
 1  y  Lm
 1  y  y i y
dy

 
This integral can be simplified: (1-y) Lmcan be approximated by the
arithmetic mean between (1-y) and (1-y)*).

y2 dy 1 1 y2
N OG  Ln
y1 y y
*
2 1  y1

Demonstration:
 1 y    1  y*  
1  y  Lm
2
1 
y y *
2

N OG
y2 dy  1 y    1 y 
*

y
1 
 1  y  y*  y  2

1 y2 dy
 1
y2  1  y  dy*

2 y1 y*  y 2 y1  1 y   y  y  *

1 y2 dy
 1
y2  1  y y y *
 dy
2 y1 y y
*
2 y1  1 y y y  *

dy dy 1 1 y2
 1
y y2 y
 Ln
y1 y y2*
2 y1 1 y y1 y y2*
2 1  y1

y2 dy 1 1 y2
NG  Ln
y1 y i y 2 1  y1

x2 dx 1 1  x1
NL  Ln
x1 x x i 2 1 x2

x dx 1 1 x2
 Ln
x1 x x I
2
2 1  x1

26
UNIT OPERATIONS COURSE ADSORPTION

x2 dx 1 1  x1
N OL  Ln
x1 x x * 2 1 x2

In terms of molar ratios:

Y2 dY 1 1  Y2
NG  Ln
Y1 Yi Y 2 1 Y 1

Y2 dY 1 1  Y2
N OG  Ln
Y 1 Y Y
*
2 1 Y 1

In the case of dilute concentrations, the equation NOGis reduced to :

y2 dy
N OG
y1 y*  y

If we additionally assume that the equilibrium and operating curves are

lines:

y* m x b
L
y x x 2   y2
G

y2 dy
N OG
y1 y*  y

y y * q x s

L x2 dx L y y *
N OG Ln 1

G x1 qx s Gq y y * 2

y1y1 y1 y 2
N OG
 
y1 y1*   y 2 y *2  y Lm
y1 y1*
Lm : logarithmic mean
Ln
y 2 y *2

If Y* = mx

R y  mx 1 1 L
N OG Ln 1 2 1  with R G slope of the lines
R 1 y 2 mx 2 R R
m
operational and balance.

27
UNIT OPERATIONS COURSE ADSORPTION

R is the absorption extraction factor.

7.2 Determination of mass transfer coefficients

G
y1
S dy
The height of the column h .
y2 ky a y y i

The correlations have been established for the calculation of the HUT.Gand HUTLpartial exams
we access the HUTO,G,Lglobal through relationships:

1 L
HUTOG HUTG HUTL
R R G
HUTOL HUTL R HUTG m

The height h = NOG. (HUT)OG= NOL. (HUT)OL

= NG. (HUT)G= NL. (HUT)L

The height of the transfer unit (HTU) is an average for the whole
columns. The relationships between the coefficients and the HUT are:

G L
 hut  OG S  HUT  OL S
K y.a K x.a

G L
 HUT  G S  HUT  L S
k y.a k x.a

We have the following relationships:

1 1 P

K y.aKy.akx.a.C

G G L
S S  S p G
K y.aKy.akx.a. C L

28
UNIT OPERATIONS COURSE ADSORPTION

G
 HUT  OG  HUT  G   HUT  L m
L
1
 HUT  OG  HUT  G   HUT  L
R
 HUT  OL  HUT  L  R  HUT  G

Kyand Kxare the global mass transfer coefficients determined

either experimentally or from empirical equations.

Sh A Re Sc

D
where Re = Reynolds number =

Schmidt number DG
KC
Sh = Sherwood number = DG

Where D diameter of the column

DG diffusivity coefficient
KCmass transfer coefficient
viscosity
µ speed

One can define the equivalent height to a theoretical plateau and the number
equivalent of theoretical plates.

Z = HEPT . NEPT = NOG.HOG

HEPT N OG logR
H OG NEPT 1
1
R
height of the column
N OG
height of transfer unit

In the specific case where the equilibrium line and the operating line are
for straight lines, the Kremser – Brown – Souders equation is applied:

y1 y 2 R n1 R
y1 m x 2 R n1 1
where n is the number of theoretical floors
R is the ratio of the slope of the operating line to the line.
of balance.

29
UNIT OPERATIONS COURSE ADSORPTION

x 1 x 2 R n 1
y1 R n1 1
 x2
m

Colburn demonstrated that it could be transformed and obtained:

x2

y2
y 1 m x 2 1 1
Ln 1 
y 2 m x 2 R R
n NEPT
LnR

y1

x1

By changing the indices:

X0

Y1

y N1 x 0 1 1
Ln 1 
y1 m x 0 R R
n NEPT
LnR
YN+1

XN

kx
The following correlations allow us to calculate the ratio.  :
ky

Pratt's equation
0.25
G G 0.67
S 1 dc S
 HUT G G

k y.a C a G G DG
4
dc
a

Sherwood Holloway Equation

30
UNIT OPERATIONS COURSE ADSORPTION

0.25
L 4L 0.5
S  HUT  L 119 S L

k x.a L
.a L DL

1
2 3
with 2
L
thickness of the liquid film.
L g

8- Efficiency of Murphree trays

In reality, on a real plate, equilibrium is not achieved. yi y *I andx i x *I

we define.

real impoverishment real enrichment

G L
maximum impoverishment maximum enrichment

ideal case
Y
YI Xi+1
Yi-1

i G
=1

L
=1
Yi=Y i
*

Yi-1 Xi
X
Xi+1 Xi=Xi*

Yi= Yi*(Xi)
Xi= Xi*(Yi)

b/ Real case

Y
Yi Xi+1
Yi-1

G
=1

Yi =1
L

Yi*
Yi-1 XI
X
31 Xi+1 XI Xi*
UNIT OPERATIONS COURSE ADSORPTION

Yi1 YI X i X i1
G L
Yi1 Y * i
X *i X i1

On a plateau, we obtain a variation in concentration of y.i-1 to yI

The real variation is yi-1-yi

The ideal variation would be yi-1-yi*

Thus the Murphree efficiency EM :

y i1 y I
EM
y i1 y *I

9- Non-isothermal absorption

Several absorbers operate with gaseous and liquid mixtures.

diluted and satisfy in these cases the assumption that the operation is isothermal.
However, the absorption operation is generally exothermic and when of
large quantities of dissolved gas are absorbed to form more solutions
concentrated, the effects of temperature can no longer be ignored. If, because of
From the absorption, the temperature of the liquid increases, the solubility of the solute can
be greatly reduced and the capacity of the absorber decreases (or a flow of
higher liquid is required). If the heat released is very great, a circuit of
Cooling must be placed inside the absorber where the liquid can be
renewed, cooled and then sent back to the absorber.

During the study of isothermal absorption, the equilibrium curve Y* = m.X

is an isotherm. If the heat of absorption is not to be overlooked, the curve
equilibrium is no longer isothermal and we will define a height Z will correspond to
an increase in concentration Z.

One can consider that the process is isothermal in Z.

X2

T0
XZ Y2
T1 à Z X T
T2

The differential heat of absorption

Y2

32

X1
UNIT OPERATIONS COURSE ADSORPTION

dT
q X 
dX
Y
Y1 Operating line

Equilibrium curve

T4

T3
2 T2

T1

T0

X
X2 X1
X

Starting from point (X2, T0we absorb X, the temperature rises by T1,
we obtain the point (X2+ X, T0+ T1=T1).
A second time we obtain the point (X2+2 X, T0+ T2=T2), ( T1 T2) car the
the variation in temperature is not linear with X. etc...

We obtain a new non-isothermal equilibrium curve and its shape is no longer

specific to the system, but depends on point X2.

Let’s consider a column where QYis the amount of heat per unit of
cleared time such that an energy balance gives:

L The.HLO G N.HGp 1 Np1

L Np.HL G
N
1 .HG Q Y 1
 p

H represents the molar enthalpy.

We refer to the base temperature T0where each substance has a

zero enthalpy, at T0and 1 atm.

The liquid enthalpyH L C L  T  T0   H S; HS being the heat

integral of solution or the molar enthalpy of mixing.

33
UNIT OPERATIONS COURSE ADSORPTION

In the case of an adiabatic process Qyis zero and the temperature at

The input of the absorber is greater than that of the output.

L N G N p1
LNG
p
n1
G1

L0 LN x N G N 1
y
N p 1
LN p x G
N p1
Yn1
1
Ln-1 Gn
The enthalpy balance:

L N.HL G
n
N.HG p 1 Np1
L Np.HL G
N
n1 .HG n1

 p

Ln Gn+1
Gp n+1
Np

L
The temperatures
N P
of currents L and
0 G are
Np+1 known. It is necessary
to estimate the temperature T1gas G1which is that of the temperature of the floor
1) and which is used in the overall balance equation to determine the
temperature of the outgoing liquidNp L.

For ideal solutions HS for a mixture of liquid is zero and

The enthalpy is equal to the sum of the enthalpies of the different constituents.

Example of non-isothermal absorption processes:

To obtain H2SO4we use the absorption of SO3by the water for

obtain H2SO4at 70%. To obtain H2SO4at 98% we use absorption of
SO3by H2SO4at 70% because it's a mixture that does not emit heat
the inverse of the mixture SO3+H2O who has a very significant effect to the point of
determine the solubility of SO3in the water.

10. Desorption

A gas dissolved in a liquid is released and enters the gas phase.

The calculation process is identical except for one difference, the line
operating is below the equilibrium line.
y y*
yi
x y2
y yi
y *

xI
xI y1

34
x
x x1
*
x x2
UNIT OPERATIONS COURSE ADSORPTION

Desorption y < y*
x > x*
XI- Case of co-current flow

Y2 Absorption
Absorption

Y2Y1
Y1 X2X1

Desorption

Y1
Desorption
Y2< Y 1
X2X1
Y2

X2 X1X1 X2

35