1

Absorption for HAP

and VOC Control

Hsin Chu
Professor
Dept. of Environmental Engineering
National Cheng Kung University
2
1. Introduction
Absorption is a diffusional mass-transfer operation.
Soluble gaseous component is removed from a gas
stream by dissolution in a solvent liquid.
The driving force for mass transfer: concentration
difference
Driving force is positive: absorption
Driving force is negative: desorption or stripping
Solvent: water or a low volatility organic liquid
Adequate solubility
3
2. Aqueous systems
Acidic gases such as HCl, HF, and SiF
4
can
be absorbed in water efficiently.
Alkaline is better.
Less soluble acidic gases such as SO
2
, Cl
2
,
and H
2
S can be absorbed more readily in a
dilute caustic solution.
Na
2
CO
3
, NaHCO
3
, or NaOH < 5~10%
CaO, Ca(OH)
2
, CaCO
3
: plugging, scaling
NH
4
OH: NH
3
slip
4
CO
2
: consume more caustic solution
pH < 9: the amount of CO
2
absorbed can be
kept low
NH
3
: can be removed with acidic water
solutions such as dilute H
2
SO
4
, H
3
PO
4
, or
HNO
3

Becomes fertilizer ingredients
5
3. Nonaqueous systems
Amines
Monoethanol-, diethanol-, and triethanolamine;
methyldiethanolamine; and dimethylanaline
For CO
2
, SO
2
and H
2
S absorbing
High absorbent cost: absorbent regeneration is a must
Organic solvent
Heavy oils, other solvents with low vapor pressure
For organic vapors absorbing
Absorber- stripper system
6
4. Types and Arrangements of Absorption Equipment
Contact between a gas and a liquid
Break the liquid up into small droplets or
thin films, which are constantly renewed
through turbulence to provide high liquid
surface area for mass transfer
Most commonly used devices
Packed and plate columns
Open spray chambers
Cyclonic spray chambers

7
Next slide (Fig. 11.1)
Countercurrent packed column: maximizes
driving force, because it brings the least
concentrated outlet gas into contact with fresh
absorbing liquid
However, it plugs rapidly when appreciable
particulates are present.
Second slide (Fig. 11.2)
Spray chamber
10
Next slide (Fig. 11.3)
Cyclonic spray chamber
Second slide (Fig. 11.4)
Plate column
Bubble cap tower: greater variations in gas flow rate
Sieve plates: simple, low cost but narrow gas flow operating
ranges
Third slide (Fig. 11.5)
Baffle tray tower: plug resistant, mass transfer is poor.
14
5. Design Techniques for Countercurrent Absorption
Columns
Co-current towers
Cross-flow towers
Countercurrent towers: the most frequently used
Two main factors:
The height: determined by the rate of mass transfer
The diameter: determined by the quantity of gas passing up the
tower
The upper limit of flow occurs when the tower begins to flood
Higher flow rates result in higher pressure losses
Assume the columns operate isothermally because of the dilute
solutions: for the following design equations
15
5.1 Equilibrium Relationships
Vapor-liquid equilibria for miscible systems
Equality of fugacity in both phases at the same temp.
and pressure:

i: component i, f: fugacity, v: vapor, L: liquid
Component i in the liquid phase:

x
i
: molar fraction of component i in the liquid,
i
:
activity coefficient, :standard state fugacity of the
pure component only
0
i
f
0

L
i i i i
f x f =

V
i
f

L
i
f =
16
Component i in the vapor phase:

y
i
: molar fraction of component i in the vapor,
: fugacity coefficient, P: pressure of the system
Therefore,

i
|

V
i i i
f y P | =
0
i i i i i
y p x f | =
17
The Gibbs-Duhem equation for n components


M: solution property, :the corresponding partial molar
property, T: temp.
Using excess Gibbs free energy changes, above equation
can be written in terms of the enthalpy and volume
changes on mixing H and V, respectively, and the
activity coefficient,
i
:

2
1
(ln )
n
i i
H V
dT dP x d
RT RT

A A
+ =

1
(ln ) 0
n
i
i
P T
M M
dT dP x d M
T P
c c
( (
+ =
( (
c c


i M
18
Solutions of nonpolar molecules or solutions at low
concentration, H and V are small
Therefore, for a binary mixture:

For a solution at low pressure where the vapor is an ideal gas:
presuming to be equivalent of the pure component fugacity.
At constant temp., above equation can be rewritten:

0
i
f
1 1 2 2
ln( ) ln( ) 0 x d y P x d y P + =
1 1 2 2
(ln ) (ln ) 0 x d x d + =
19
, the partial pressure of component 1


Therefore,

Integrate,
Therefore, from partial pressure data of one
component of a binary solution, the partial pressure of
the second component may be calculated.
1 1
y P P =
2 2
y P P =
2 1
1 x x =
2 1
0
2
2 1
0
(ln ) (ln )
P P
P
d P d P =
} }
2
2 1 1
0
0
1 1
2
ln
(1 )
P
P x dP
x P
P
(
= (

(

}
20
5.2 Ideal Solutions-Henrys Law
A general relationship for fugacity of a component in
a liquid mixture in terms of the pure component
fugacity:

(1)
An ideal solution:
Therefore, , called the Lewis and
Randall Fugacity Rule
id: ideal solution
0

1
ln ( )
L
p
i
i
i L
i i
f
v v dp
RT x f
(
=
(
(

}
0
i i
v v =

L
L id
i i i i
f f x f = =
21
More generally to be consistent with the
previous equation

Next slide (Fig. 11.6)
Broken lines: ideal solution, : the fugacity
of a pure component as it actually exists,
: an imaginary state of the real solution
Solid line: a real solution of Eq. (1)

0 id
i i i
f x f =
0
( )
i
f A
0
( )
i
f B
23

For a pure component
Henrys Law


For a dilute solution
Rewrite:

k
i
: Henrys Law constant
0
1.0

( )
lim
i
L
i
i i
x
i
f
f A f
x

(
= =
(
(

0
1.0

( )
lim
i
L
i
i
x
i
f
k f B
x

(
= =
(
(

L
i i i
f x k =
24
From previous equation:

For solutions at low pressure,
and (partial pressure)
Therefore, (common form of
Henrys Law)

i i i i
y P x k | =

1.0
i i
| | ~ ~
i i
y P P =
i i i
P x k =
25
Rewrite:


Define m=k
i
/P (a common form for Henrys Law)
Therefore, y
i
= mx
i
(2)
Henrys Law is an adequate method of representing
vapor-liquid phase equilibria for many dilute solutions
encountered in absorption work.
However, weak electrolytes: dissociation of the
electrolyte may take place upsetting the normal
equilibrium relationship.
i
i i
k
y x
P
(
=
(

26
5.3 Countercurrent Absorption Tower Design
Equations
Next slide (Fig. 11.7)
A packed tower with countercurrent flow
Mass balance around the cross-section A-A:
y
A
G + x
A
L = (y
A
- dy
A
)G + (x
A
+ dx
A
)L
Therefore, G dy
A
= L dx
A

However, the flow rates do not remain constant through the
column:
d(Gy
A
) = d(Lx
A
)
28
Integrate from the top of the column down to the
cross-section A-A:
y
A
G - y
A1
G
1
= x
A
L - x
A1
L
1
(3)
Rewrite the flow rates on a solute-free basis (use B):

B A A1 A1
1 1
G = (1 y )G = (1 y )G
(1 ) (1 )
B A A A
L x L x L

= =
29
Rewrite Eq. (3):

(4)
For dilute solution:



Eq. (4) reduces to Eq. (3) which is called the operating line.
Eq. (4): general equation with a general equilibrium
relationship
Eq. (2): Henrys Law Equation
, called the equilibrium line (will be shown later)

1 1
1 1
1 1 1 1
A A A A
B B B B
A A A A
y x x y
G L L G
y x x y
( ( ( (
+ = +
( ( ( (


1 1
,
B A B A
G G G L L L ~ ~ ~ ~
*
A A
y mx
-
=
1
(1 ) 1.0, (1 ) 1.0, .
A A
y x etc ~ ~
30
5.4 Origin of Volume-Based Mass-Transfer
Coefficients
Several theories have been devised
Whitman two-film concept remains an excellent basis
Molecular diffusion serves a basis for mass transfer
coefficients
Driving force across the interface: concentration gradient
Complete thermodynamic equilibrium: no concentration,
temperature, or pressure difference in the areas of the diffusion
5.4.1 Steady-State Molecular Diffusion
31
Assume that only one component is in motion across the
interface
Molecular diffusivity: a relative velocity of the diffusion
component with respect to an average velocity of the entire
stream is a direction normal to the interface
The molecular motion is accounted for by Ficks Law.
Total flux is the sum of the molecular diffusion and the
convective bulk flow.
32
: the flux of component A in a mixture with component B
(unit: moles/time/area)
And
Component A with mole fraction, y
A
Molecular diffusion flux in the Z direction =
where, D
AB
is the molecular diffusivity and is the density of
the gas

A
N
bulk A B
N N N = +
A
AB m
dy
D
dz

m

( )
A
A A B
A AB m
dy
N y N N D
dz
= +
33
Absorption of a single component A: it is assumed that A is
diffusing through a stagnant layer of B, thus .
Previous eq.: can be rearranged and integrated
over a distance L with the following boundary conditions,
y
A
= y
A0
, when z = 0
y
A
= y
AL
, when z = L
The result is

(5)

0 B N =
A A
Gdy Ldx =
0
1
ln
1
AB m AL
A
A
D y
N
L y
(
(
=
(
(



34
Because y
A0
+y
B0
= 1
y
AL
+y
BL
= 1
A loghrithmic mean concentration can be difined:

(6)
A mass transfer coefficient can be defined:
(7)
Eq. (6) & Eq. (7) can be substituted into Eq. (5):

0 0
0
0
( ) ( )
ln( / )
1
ln
1
B BL A AL
BM
B BL
AL
A
y y y y
y
y y
y
y

= =
(
(

AB m
my
BM
D
k
Ly

=
0
( )
A my A AL
N k y y =
35
Next slide (Fig. 11.8)
A schematic diagram

5.5 The Whitman Two-Film Theory
37
The theory summarized:
(1) The gas film and the liquid film, on either side of the interface
(2) Transfer in the bulk phases by convection currents
Concentration differences are negligible except in the vicinity of
the interface
(3) On each side of the interface, convection currents die out
(4) Both films offer resistance to mass transfer
Transfer takes place through these films by a mechanism similar
to molecular diffusion
(5) The interface is at equilibrium and offers no resistance to mass
transfer
(6) Uniform composition in the main stream
(7) At the interface, y
Ai
and x
Ai
are in equilibrium described by
Henrys Law y
Ai
= mx
Ai

38
Packed Column
The fluid is in turbulent motion
Mass transfer through the films is defined by k
y
and
k
x
(turbulent mass transfer coefficient)

( ) ( )
A y A Ai x Ai A
N k y y k x x = =
39
5.6 Overall Mass-Transfer Coefficients
Mass-transfer coefficients defined on the basis of
interfacial mole fractions have little practical value
since these mole fractions cannot be measured.
Two new pseudo-mole fractions are defined as
illustrated in the next slide (Fig. 11.9)
Overall mass transfer coefficients K
y
and K
x
are
defined based on these new pseudo-mole fractions.
y
A
*
= equilibrium mole fraction of the solute in the
vapor corresponding to the mole fraction x
A
in
the liquid
x
A
*
= equilibrium mole fraction of the solute in the
liquid corresponding to the mole fraction y
A
in
the vapor
41
The mass transfer may now be written:

The construction in Fig. 11.9 (last page) illustrates
how to determine x
A
*
for any y
A
, and y
A
*
for any x
A
.
* *
( ) ( )
A y A A x A A
N K y y K x x = =
42
The overall coefficients (K
y
& K
x
) can be related to
the individual phase coefficients (k
y
& k
x
) making
use of Henrys Law.
Therefore:


Also:

*
( )
A y A Ai Ai A
N K y y y y = +
* *
( ) ( ) ( ) ( )
x
A y A A y A Ai x Ai A Ai A
k
N K y y k y y k x x y y
m
= = = =
43
Combining:

Rearranging:

[ ]
A A
A y
y x
N mN
N K
k k
= +
1 1
y y x
m
K k k
= +
44
5.7 Volume-Based Mass-Transfer
Coefficients
Packing has a characteristic area per unit volume of
packing:


Flow rate of vapor and liquid phases, viscosity,
density, surface tension, and size and shape of the
packing determine the value of a.
These same factors affect the value of the mass
transfer coefficients K
y
and K
x
.
interfacial area
a
unit volume
=
45
Therefore, two new quantities K
y
a and K
x
a
(volumetric mass transfer coefficients) are common
to be used.
If A is the absorption tower cross-sectional area, and
Z the packing height, then Az is the tower packing
volume.
Defining A
i
as the total interfacial area:
A
i
= aAz
or in differential form:
dA
i
= aAdz
46
Rewrite in dN
A
, with units (moles/unit time) =
(moles/unit timearea) (area):



or dN
A
= K
y
a(y
A
-y
A
*
)Adz = K
x
a(x
A
-x
A
*
)Adz


*
( )
A A i y A A
dN N dA K y y aAdz = =
A
N
47
5.8 Determining Height of Packing in the Tower:
the HTU Method
For the gas phase, the differential rate of mass
transfer of component A is equal to the differential
rate of change of the mass of A in the incoming gas
stream.
dN
A
= d (Gy
A
)
Rewrite the flow rate on a solute-free basis:
G
B
= (1-y
A
)G, where G
B
is a constant

then,


( )
(1 ) (1 )
B A A
A
A A
G y Gdy
d Gy d
y y
( (
= =
( (


48
Therefore,

(8)
A similar equation based on the liquid phase can be
written which would be useful in stripping
calculations.


*
( )
(1 )
A
y A A
A
Gdy
K a y y Adz
y
=

49
Define a mass transfer coefficient which is
independent of concentration:



where (1-y
A
)
LM
is the log mean concentration defined
by:


*
*
(1 ) (1 )
(1 )
(1 )
ln
(1 )
A A
A LM
A
A
y y
y
y
y

=
(

(1 )
o
y
y
A LM
K
K
y
=

50
Define flow rates based on tower cross section:



Rewrite Eq. (8):



*
( )
1 (1 )
o
y
A
A A
A A LM
K a
dy
G y y dz
y y
(
=
(


G
G
A
=
L
L
A
=
51
Assuming constant flow rate, , through the
column, this equation can now be integrated from z =
0 to z = z down the column to determine tower height:

0
1
*
(1 )
(1 )( )
A
A
y
A LM
A
o
y
y A A A
G y
Z dy
K a y y y
(

=
(

(

}
G
52
The ratio of flow rate to mass transfer has been
designated as the height of a transfer unit (HTU), or,
for the gas phase, H
OG



Therefore, H
OG
has been defined in such a way that it
remains constant through the absorption column.

0
OG
y
G
H
K a
(
=
(
(

53
The number of overall mass transfer units (NTU), or
for the gas phase, N
OG
has been defined:


The height of the column may now be calculated
from:
Z = H
OG
N
OG
0
1
*
(1 )
(1 )( )
A
A
y
A LM
OG A
y
A A A
y
N dy
y y y

=

}
54
5.9 Dilute Solution Case
For dilute solutions, Henrys Law is usually a good
choice for an equilibrium relationship.
y
A
*
= mx
A
(9)
(1-y
A
) (1-y
A
)
LM
1.0
Therefore,

(10)
The operating line, y
A
G - y
A1
G
1
= x
A
L - x
A1
L
1
, may
be rewritten as:

(11)
0 0
1 1
* *
(1 )
(1 )( ) ( )
A A
A A
y y
A LM A
OG A
y y
A A A A A
y dy
N dy
y y y y y

= =

} }
1 1 1 1
( )( ) ( )( )
A A A A A A A
L L
y x x y mx mx y
G mG
= + = +
55
Note that
The absorption factor, Ab, can be defined as:
(12)
Use Eqs. (9), (11), and (12):

(13)
Substitute Eq. (13) in Eq. (10) and integrate:


When a pure solvent such as water is used, x
A1
= 0.0:
/ / Ab L mG L mG = =
*
1
1
A A
A A
y y
y mx
Ab

(
= +
(

{ }
0 1 1 1
ln [( ) /( )](1 1/ ) (1/ )
(1 1/ )
A A A A
OG
y mx y mx Ab Ab
N
Ab
+
=

( )( )
0 1
1
ln / 1 1/ 1/
1 (1/ )
OG A A
N y y Ab Ab
Ab
(
= + (
(

56
6. Countercurrent Flow Packed
Absorption Tower Design
6.1 General Considerations
When a packed tower is designed for treating a
given quantity of gas per hour, the height of
packing is determined from mass transfer
consideration, and the diameter or cross-sectional
area is determined by the gas velocity in the
empty tower cross-section.
Next slide (Table 11.1)
Economic factors in packed tower design

58
6.2 Operations as Packed Towers
In a packed tower operating in counterflow at
constant liquid rate, the pressure drop varies
with the gas mass velocity.
With liquid holdup in the packing, there is
smaller void space in the column, and a higher
pressure drop exists than with dry packing.
Next slide (Fig. 11.10)
Pressure drop vs. superficial gas mass velocity
L: liquid flow rate

60
From Fig. 11.10 (last slide)
The point at which the rate of change of pressure drop
increases more rapidly than a constant value has been called
the upper loading point.
The liquid holdup is now increasing with the increasing gas
rate.
The gas velocity corresponding to this upper limit is known as
the flooding velocity.
At gas rates greater than the flooding velocity, the column will
act as a gas bubbler.
The greater the liquid rate, the lower the gas velocity at which
flooding occurs.
Larger, more open packings flood at high velocities than
smaller, more dense packings.
61
Two types of liquid holdup:
(1) Static holdup is the volume of liquid per
volume of packing that remains after gas and
liquid flows are stopped and the bed has drained.
Static holdup depends on packing surface
characteristics.
(2) Operating holdup is the volume of liquid per
volume of packing that drains out of the bed after
gas and liquid flows have been stopped.
Gas flow rate has little effect on holdup below upper
loading point. Packed towers may be operated above
90% of the flooding velocity.
62
Some operational characteristics for countercurrent
packed tower:
(1) Pressure drop between 0.25 in. and 0.60 in. H
2
O
per foot of packing
(2) Air velocity between 5.0 and 8.0 ft/s with
modern high capacity plastic packing
(3) Inlet concentrations of contaminant in the gas
stream not to exceed 5,000 ppmv
(4) Liquid irrigation rates typically between 2 and 8
gpm/ft
2
of column cross-sectional area
(5) L/G: 15-90 L/m
3
(5-10 L/m
3
for FGD, 1-2 L/m
3

for particulates)
(based on (2) & (4))
63
6.3 Choosing a Tower Packing
Two major types of packing:
(1) Random dumped pieces
(2) Structured modular forms
It should have enough mechanical strength to
carry the load and allow easy handling and
installation.
It should be able to resist thermal shock and
must be chemically resistant to the fluids being
processed.
64
6.3.1 Dumped Packings
Two of the most popular shapes are rings and saddles.
Sizes range from in. to 3.5 in. with 1 in. being a very
common size.
Packings are ceramic, metal, or plastic, dependent upon the
service.
Ceramic materials will withstand corrosion and are therefore
used where the resultant solutions are aqueous and corrosive.
Metals are used where nocorrosive organic liquids are present.
Plastic packing may be used in the case of corrosive aqueous
solutions and for organic liquids which are not solvents for the
plastic of which the packing is made.
Next slide (Fig. 9-34)
Common packing materials
66
Metal packing is more expensive, but provides lower
pressure drop and higher efficiency.
When using plastic materials, care must be taken that
the temperature is not too high and that oxidizing
agents are not present.
Ring-type packings are commonly made of metal or
plastic except for Raschig rings, which are generally
ceramic.
Ring-type packings lend themselves to distillation
because of their good turndown properties and
availability in metals of all types that can be press
formed.
67
Usually ring-type packings are used in
handling organic solutions when there are no
corrosive problems.
However, rings do not promote redistribution
of liquids, and Raschig rings may even cause
maldistribution.
Saddles are commonly made from ceramic or
plastic and give good corrosion resistance.
Saddles are best for redistribution of liquid.
68
6.4 Choosing a Liquid-Gas Flow Ratio
The operating line, y
A
G - y
A1
G
1
= x
A
L - x
A1
L
1
, may
be solved for the exit liquid concentration x
A0
:


Therefore, if L or G/L can be found, the entire
material balance will be solved.
Next slide (Fig. 11.11)
A plot of y
A
vs. x
A
0 0 1 1
( )
A A A A
G
x y y x
L
| |
= +
|
\ .
70
Last slide Fig. 11.11 (left panel)
With a concave equilibrium line, three possible
operating lines are shown. L/G is the slope of the
operating line.
The line furthest to the right with the smallest slope
represents the limit to which the slope of the
operating line can be drawn.
Thus, this slope represents a minimum L/G ratio in
which x
A0
and y
A0
are actually in equilibrium, and the
line could not be drawn with a smaller slope because
it would go through equilibrium.
71
Since true thermodynamic equilibrium can
never be reached practically, this point
represents a theoretical minimum L/G which
never can be attained in practice.
Real L/G ratios are usually set at 1.1 to 1.7
times this minimum ratio.
72
Fig. 11.11(right panel)
With S-shaped equilibrium curve, the
minimum value of L/G is reached at some
point within the column where the operating
and equilibrium lines are tangential and
thermodynamic equilibrium would be
exceeded if the L/G ratio were any smaller.
73
6.5 Determining Tower Diameter-
Random Dumped Packing
Next slide (Fig. 11.12)
Use a modified Sherwood flooding correlation
to determine tower diameter
Second slide (Table 11.2)
The definition of terms used in Fig. 11.12

76
The packing factor, F, is a characteristic of a
particular packing size and shape.
Oringinally it was defined as a/
3
where a is the
interfacial area per packed volume and is the void
fraction.
However, this definition for F did not adequately
predict the packing hydraulic performance.
Therefore, Table 11.3 (next slide) presents packing
factors which have been calculated from
experimentally determined pressure drop.
78
How to use Fig. 11.12
From the abscissa, the packing factor, and the
required pressure drop per foot of packing, the
ordinate of the curve can be read and the value of G
*

in lb/ft
2
s calculated.
With G converted to lb/s the column cross-sectional
area A
s
is then:

The column diameter can be found from:
2
* 2
( / )
( )
( / )
s
G lb s
A ft
G lb s ft
=
-
*
4G
D
G t
=
79
6.6 Determining Tower Diameter-
Structured Packing
Sturctured packings are being favored because they have a
higher capacity than dumped packing and scaleup is more
predicable.
Next slide (Fig. 11.13)
A capacity correlation for nonfoaming systems
Flow parameter,
Efficient capacity C
SC
(ft/s) =
Where

L
= liquid surface tension in dyne/cm,
5
L
73 dyne/cm

L
= liquid viscosity in cp, 0.07
L
1.1 cp
Design is carried out routinely up to 90% of efficient capacity.
G
L
L
X
G

=
0.16 0.11
0
20 0.2
L L
C
o

( (
( (


81
Because the pressure drop curves for random and structured
packings are similar, pressure drop for structured packing can
be estimated from the generalized correlation of Fig. 11.12.
However, in this case the packing factor is affected by the
operating pressure.
Useful packing factors from Strigle are reported in Table 11.4
(next slide) for five different sizes of Intalox Structured
packing.
Table 11.4:
The smaller packing factor should be used for pressure below
10.0 psia or at liquid rates below 4 gpm/ft
2
.

83
6.7 Controlling Film Concept
The Whitman Two-Film Theory
The film that offers the predominant mass
transfer resistance is called the controlling film.
84
The overall mass-transfer coefficient can be solved:



Gases with a limit solubility in the liquid phase have a high
value of m.
Therefore, above eq. reduces K
y
k
x
/m liquid film
controlling
Gases with a high solubility in the liquid phase have a low
value of m.
Therefore, above eq. reduces K
y
k
y
gas film controlling
1
1
y x
y
x y
y x
k k
K
m
k mk
k k
= =
+
+
85
6.8 A Correlation for the Effect of L/G Ratio
on the Packing Height
The correlation is based on a previous Eq.:

(14)
Henrys Law is a straight equilibrium line, and the
correlation presented in this section assumes
Henrys Law for the equilibrium relationship.
| |
0 1
1
ln ( / )(1 1/ ) 1/ )
1 1/
OG A A
N y y Ab Ab
Ab
(
= +
(

86
At the minimum L/G ratio y
A0
= mx
A0

Eq. can be solved for L/G:


For any y
A
there is an x
A
determined by the operating
line:


There is also a in equilibrium with the x
A
such that
/m.
1 1
( )
A A A A
L
y mx mx y
mG
| |
= +
|
\ .
0 1 1 0
1 0
min 0 0 0
1 /
(1 / )
/
A A A A
A A
A A A
y y y y L
y y m
G x x y

| |
= = =
|
\ .
1 1
( )
A A A A
L
y x x y
G
| |
= +
|
\ .
*
A
y
*
A
y
87
Assuming that fresh solvent is being used x
A1
= 0.0
and the operating line may be rewritten:


Therefore,

(15)
Define

(16)
which is the same as before
*
1
A
A A
y L
y y
m G
| |
= +
|
\ .
*
1
( )
/
A A
A
m y y
y
L G

=
( / ) L G
Ab
m
=
88
Define B as a multiple of (L/G)
min
to set the actual (L/G):
(L/G) = B(L/G)
min

With
Then,
Substitue Eq. (16) into Eq. (15) (previous slide):

For the dilute case, substituting this into:
1 0
min
(1 / )
A A
L
y y m
G
| |
=
|
\ .
min
1 0
( / )
(1 / )
A A
B L G
Ab B y y
m
= =
*
1
1
( )
A A A
y y y
Ab
| |
=
|
\ .
0
1
*
( )
A
A
y
A
OG
y
A A
dy
N
y y
=

}
89
Integrating N
OG
:


This is the same as Eq. (14).
Recall Ab is defined as a function of B:
Ab = B(1-y
A1
/y
A0
)
If the value of (y
A1
/y
A0
)is fixed at 0.90, 0.95, and 0.99,
N
OG
can be plotted vs. B with (y
A1
/y
A0
) as a parameter
(next slide, Fig. 11.14).
| |
0 1
1
ln ( / )(1 1/ ) 1/
1 1/
OG A A
N y y Ab Ab
Ab
(
= +
(

91
Fig. 11.14 (previous slide)
For any given B and percent removal, knowing
and a, H
OG
can be calculated, and the height of the
column can be calculated from Z = H
OG
N
OG
.
Since Eq. (14) is derived for the dilute solution,
examination of Fig. 11.14 shows that the number of
transfer unit, N
OG
, cannot be greatly reduced by
increasing B, the multiple of the minimum L/G ratio,
above 1.70.
G
0
y
K
92
6.9 Henrys Law Constants and Mass-
Transfer Information
Henrys Law constants can be found in the
chemical Engineers Handbook.
It is also possible to calculate Henrys constant
for solubility data, seen in the example in
Table 11.5 (next slide).
94
The measurement of mass-transfer coefficients has
occupied many experimenters of the past 70 years.
It is surprising that there is not much data available in
the open literature.
The problems of these measurements:
(1) Difficulty in determining the interfacial area
(2) Difficulty in determining individual coefficients
from measurements of overall coefficients
(3) Entrance effects

95
The packing area becomes effective only if it is
wetted.
Thus, K
y
a must include a hidden factor which reflects
the proportion of the packing area that is effective.
This effective area is proportional to the follwing
factors:
(1) Liquid rate
(2) Type of packing
(3) Method of packing the tower
(4) Nature of the liquid distribution
(5) Height and diameter of the tower
96
Next two slides (Table 11.6)
Mass-Transfer Coefficient data
Mass-Transfer Coefficient for structured
packing are generally greater than dumped
packing, indicating improved mass transfer.

99
6.10 Using Henrys Law for
Multicomponent Solutions
For a binary solvent composed of j and k pure
components, Carroll suggests:


Where H
i, mix
: the Henrys constant for solute i in a
mixture of j and k
H
ij
, H
ik
: the Henrys constants for the pure
solvents
x
j
, x
k
: the mole factions of the solvents
a
jk
: an interaction paramter that must be
obtained from experimental data
,
ln( ) ln( ) ln( )
i mix j i k ik jk j k
j
H H H x H a x x = + +
100
In the absence of any other information it may be
assumed that a
jk
in the previous Eq. is independent of
i or set equal to zero.
For mixed solutes at low pressure, it may be assumed
that each gas is independent of the other.
Therefore, for two solutes i and j:
x
i
H
ik
= y
i
P
x
j
H
jk
= y
j
P
Thus the solubility of the individual components is
determined by its own partial pressure as if the other
component were not present.
101
7. Sample Design Calculation
To remove a dilute concentration VOC from an air
emission with a dilute absorbing solution
The problem
An air stream from a electronic circuitry
manufacturer contains 50 ppm of a VOC which must
be treated.
Q = 10,000 acfm
P = 1 atm
T = 20
102
Emission standard = 90% removal (5 ppm)
Absorbing liquid = water
Henrys Law constant, m = 2.0
Mass-transfer coefficient = 4.0 lb moles/hft
3
L/G = 1.5 (L/G)
min

Packing = 1 ceramic Intalox saddles
P = 0.5 H
2
O/ft of packing
Packing factor, F = 92
MW (air) = 29.0

103
The physical properties of the water:
MW = 18
T = 20
P = 1 atm
Density = 62.3 lbm/ft
3

Viscosity = 1.0 cp
Kinematic viscosity = 1.0 cst
Design the absorption column
104
Solution:
m = 2.0
y
i
= 2.0x
i

The operating line based on :
y
A
= (L/G)(x
A
-x
A1
) + y
A1

Minimum L/G ratio can be calculated by:


y
A0
= 0.000050 (50 ppm) y
A1
= 0.000005 (5 ppm) x
A1
= 0.0
0 1
0 1
( )
( / )
( )
A A
A A
y y
L G
x x

* 0
0 0
0.000050
0.000025
2.0
A
A A
y
X x
m
= = = =
min
0.000050 0.000005
( / ) 1.8
0.000025
L G

= =
105
The minimum slope operating line is
y
A
= 1.80 x
A
+ 0.000005
Operating (L/G) = 1.5 (L/G)
min
= 1.5 1.80 = 2.70
L/G = can be solved for the operating
x
A0
with x
A1
= 0.0


The operating line then becomes
y
A
= 2.70x
A
+ 0.000005
0 1
0 1
A A
A A
y y
x x

0 1
0
0.000050 0.000005
0.0000167
2.70
( )
A A
A
y y
x
L
G

= = =
106
The tower diameter can now be calculated.




Calculate the superficial velocity from Fig. 11.12
(previous):
The abscissa is
2
2 2 2
18
2.70 1.6759
29
lbs H O
L lb moles H O lbs H O lb moles H O
lbs air
G lb moles air lbs air
lb moles air
= =
0.5
0.5
0.0753
1.6759 0.058
62.3
G
L
L
G

| |
| | | |
= =
| | |
\ . \ .
\ .
107
From Fig. 11.12, read the ordinate value = 1.33
C
s
F
0.5

0.05
= 1.33


Calculate the superficial velocity V:

0.5 0.05
1.33
0.1387
92 1.0
s
C = =

0.5 0.5
0.1387
3.986 /
0.0753
62.3 0.0753
s
G
L G
C
V ft s

= = =
| | | |
|
|

\ .
\ .
108
Calculate the superficial gas velocity:

Calculate the mass flow rate:

Calculate the tower cross-sectional area for a ciraular
tower:

or
2
60 min/ 18.01
min
G
lbs
G V s
ft
= =
0.0753 10, 000 753 / min
G
Q lbs = = G=
2
753
41.81
18.01
s
G
A ft
G
= = =
3
2
10, 000 / min
41.81
3.986 / 60 / min
s
Q ft
A ft
V ft s s
= = =

109
Column diameter:


Calculate the height of a transfer unit (H
OG
):
2 2
2
3
18.01 60 min/ 1.0 / 29 37.25
min
37.25
9.31
4.0
OG
y
lbm lb moles
G h lb mole lbm
ft h ft
lb moles
G h ft
H ft
lb moles
K a
h ft
= =
= = =
4 4 41.81
7.30
s
A
D ft
t t

= = =
110
Calculate the number of transfer units (N
OG
):


Because x
A1
= 0.0
( )
/
2.70
1.35
2.00
L G
Ab
m
= = =
( )( )
( )( )
0 1
1
ln / 1 1/ 1/
1 1/
1
ln 0.000050/ 0.000005 1 1/1.35 1/1.35
1 1/1.35
4.64
OG A A
N y y Ab Ab
Ab
(
= + (

(

(
= + (

(

=
111
Calculate the height of the tower, Z:
Z = H
OG
N
OG
= 9.31 4.64 = 43.21 ft
or from Fig. 11.14 (previous) with B = 1.5 and
y
A1
/y
A0
= 90%:
Read N
OG
= 4.6
Z = 9.31 4.6 = 42.84 ft
Calculate the pressure drop through the column:
P = 0.5 H
2
O/ft of packing 43.21 ft
= 21.61 H
2
O
112
Calculate the liquid flow rate:






The irrigation rate is 151.7/41.81 = 3.62 gpm/ft
2
of
tower area.
Suggested rates are between 2 and 8 gpm/ft
2
. OK!
( )
2
2
2
2
2
2
/
min
1.6759 18.01 30.18
min
30.18 41.81
min
151.7
8.32 /
L
lb H O lbm air
L L G G
lb air ft
lbm
ft
lbm
ft
ft
Q gpm
lb H O gal
=
= =

= =
113
7.1 Flooding
From Fig. 11.12 (previous), the abscissa is the same,
0.058, and the flooding ordinate is then 2.40.
C
s
F
0.5

0.05
= 2.40


Calculate the superficial velocity:

0.5 0.05
2.40
0.2502
92 1.0
s
C = =

0.5 0.5
0.2502
7.1929 /
0.0753
62.3 0.0753
s
G
L G
C
V ft s

= = =
( | |
|
(

\ .

114
Calculate the gas flow rate:
Q = VA
s
= 7.1929 ft/s 41.21 ft
2
60 s/min =
18,045 acfm
Actual operation is at 10,000 acfm, therefore,
operation is at:
10, 000
100% 55%
18, 045
of flooding
| |
=
|
\ .
115
7.2 Structured Packing
To treat additional sources or to meet higher
electronic circuit production demand, consider
the possibility of increasing the air flow to
20,000 acfm by switching to structured
packing.
Choose Intalox structured packing 2T. The X
factor has already been calculated to be 0.058.
116
The same column will be used, the cross-
sectional area will remain the same.
Thus, the superficial velocity, V, and capacity
factor, C
s
, will double to 7.97 ft/s and 0.2773
ft/s, respectively.
P = 1 atm, the packing factor from Table 11.4
(previous) will be 20.
117
Calculate the ordinate of Fig. 11.12:
C
s
F
0.5

0.05
= 0.2773 20
0.5
1.0
0.05
= 1.2406
The abscissa remains the same, 0.058, from Fig.
11.12, the pressure drop is estimated to be 0.45
H
2
O/ft of packing.
The tower capacity is now determined from Fig.
11.13.
The X value of 0.058 is the abscissa of Fig. 11.13.
The ordinate value is estimated to be C
0
= 0.44 ft/s
118
The surface tension, , of water at 70F is 70 dynes/ft
2
.
The viscosity, , is 0.81 cp.


The actual capacity factor C
s
= 0.2773 ft/s is below
90% of the maximum tower efficient capacity,
0.90 C
sc
= 0.414 ft/s.
Therefore, the tower, if changed to structured packing,
could take even a greater air flow.

0.16 0.11
0
0.46 /
20 0.20
sc
C C ft s
o

| | | |
= =
| |
\ . \ .