Chapter 8: Absorption and Stripping

In addition to the distillation, there are other unit operations used for separating substances

Absorption is the unit operation in which one or more components of a gas stream are removed from the gas mixture by being absorbed onto a non-volatile liquid (called a solvent)

In this case, the solvent is the separating agent

Stripping is the operation that is in the opposite direction to the absorption, in which one or more gaseous components in a liquid stream is removed from the gas-liquid solution by being vaporised into an insoluble gas stream

In the stripping operation, the insoluble gas stream is the separating agent

What is the separating agent in the distillation?

Absorption can be physical, when the solute is dissolved into the solvent because it has higher solubility in the solvent than other gases chemical, when the solute reacts with the solvent, and the resulting product still remains in the solvent Normally, a reversible reaction between the solute and the solvent is preferred, in order for the solvent to be regenerated (

)
Similar to the distillation, both absorption and stripping are operated as equilibrium stage operations, in which liquid and vapour are in contact with each other
3

However, in both absorption and stripping operations, the columns are simpler than the distillation column; normally, they do not need condensers and re-boilers as per the distillation An example of the uses of absorption and stripping is a gas treatment plant, shown in Figure 8.1 The mixture of the gases (i.e. the gas to be treated in Figure 8.1), from which one component (e.g., CO2 or H2S) in the gas mixture needs to be removed, is passed through an absorber, in which a liquid-phase solvent (e.g., MEA or amine solvent in water) is passed though, to absorb that gaseous component
4

Figure 8.1: A gas treatment plant

(from Separation Process Engineering by Wankat, 2007)

The removal component is mixed (physically or chemically) with the solvent

The resulting solvent is heated by a heater and becomes the saturated solvent (so that the resulting solvent can be stripped)

The resulting saturated solvent is, subsequently, passed through a stripper (or a stripping column), in which the gaseous component in the saturated solvent is stripped off by a stripping gas (e.g., steam)

In this Example (Figure 8.1), the desired product for an absorber is a treated gas stream for a stripper is a recycle solvent stream

8.1 Absorption and Stripping Equilibria Absorption and stripping involves, at least, 3 components (as illustrated recently) and 2 phases Generally, for simplicity, we often assume the following: Carrier gas is insoluble (into a solvent) Solvent is non-volatile (thus, the loss of solvent due to vaporisation is negligible) The system is isothermal (constant temperature) and isobaric (constant pressure) Since the absorption or stripping system comprises 3 species and 2 phases, the degree of freedom (F ), calculated using the Gibbs phase rule, is
7

F =C -P +2 = 3-2 +2 = 3

As we assume that the system is isothermal, which results in the fact that the systems temperature is constant isobaric, which results in the fact that the systems pressure is constant the degree of freedom (F ) is reduced to one (1)

Normally, an equilibrium data is used as the

remaining degree of freedom (F )

In general, the amount of solute (i.e. a gas to be removed from the gas mixture) is relatively low
8

For a low concentration of a solute, a Henrys law is employed to express the equilibrium between the concentration (e.g., mole fraction or percent) of the solute in the gas phase and that in the liquid phase as follows
PB = H B x B

(8.1)

where
PB = partial pressure of the solute B in the

gas phase (or in the gas mixture)

x B = concentration (in mole fraction) of the

solute B in the liquid phase

H B = Henrys constant for the solute B

A mole fraction of the solute B in the gas phase (yB ) can be described by the following equation:
9

yB =

PB Ptotal

(8.2)

which can be re-arranged to

PB = yB Ptotal

(8.3)

Combining Eq. 8.3 with Eq. 8.1 and re-arranging the resulting equation gives

yB =

HB Ptotal

xB

(8.4)

As we have assumed previously that the system is isobaric [which results in the fact the systems pressure (Ptotal ) is constant], Eq. 8.4, which is an equilibrium equation for absorption and

stripping operations, is a linear relationship with

the slope of HB Ptotal
10

8.2 Operating Lines for Absorption

We have learned previously from the distillation operation that if the operating line is linear, it would be easy to use

To have a linear or straight operating line, it is required that the energy balance is automatically satisfied both liquid and gas (or vapour) flow rates are constant

In order for the energy balance to be automa-

tically satisfied, we have to assume that

11

 the heat of absorption is negligible the operation is isothermal a solvent is non-volatile a carrier gas is insoluble (into the solvent) For any gas absorber, as illustrated in Figure 8.2, by employing the above assumptions, we obtain the facts that
L1 = L2 = ... = LN = L GN = GN -1 = ... = G1 = G

(8.5) (8.6)

It is important to note that, to make the liquid and gas flow rates constant, especially for the

concentrated mixture (of either liquid or gas

phase), we cannot use the overall or total flow rates of gas and liquid
12

Figure 8.2: A gas absorption column

(from Separation Process Engineering by Wankat, 2007)

In other words, the liquid flow rate (L ) must be the molar or mass flow rate of a non-volatile solvent the gas flow rate (G ) must be the molar or

mass flow rate of an insoluble carrier gas

13

Accordingly, the mole fractions of solute B in the gas (yB ) and the liquid (x B ) phases, which are defined as moles of solute B (in the gas phase) yB = total moles of a gas mixture or moles of solute B (in the gas phase) yB = moles of carrier gas A + moles of solute B and xB = or moles of solute B (in the liquid phase) xB = moles of a solvent + moles of solute B have to be modified to
14

moles of solute B (in the liquid phase) total moles of a solution

moles of solute B YB = moles of pure carrier gas A moles of solute B XB = moles of pure solvent The relationships between YB and yB and between X B and x B can be written as follows

YB = XB =

yB
1 - yB xB 1 - xB

(8.7) (8.8)

Thus, by performing material balances for the given envelope in Figure 8.2, we obtain the following equation:

GYj +1 + LXo = GY1 + LX j

(8.9)

15

Re-arranging Eq. 8.9 for Yj +1 results in

Yj +1

L L = X j + Y1 - Xo G G

(8.10)

Eq. 8.10 is an operating line for absorption, with

L the slope of G
the Y-intercept of Y1 -

L Xo G

In order to obtain the number of equilibrium stages required for the absorption operation from the initial concentration of Xo to the final concentration of X N , the following procedure, which is similar to that for the distillation operation, is employed
16

1) Draw an equilibrium line on the X-Y coordinate 2) Draw an operating line; the values of Xo ,

L YN +1 , Y1 , and are generally known G

note that the point Xo , Y1 must be on

the operating line 3) Locate the point Xo , Y1 and step off

stages between the operating line and the equilibrium line from X = Xo until it reaches the final concentration of X N This X-Y diagram, which consists of the operating and equilibrium lines for the absorption operation, is still called the McCabe-Thiele diagram
17

An example of the McCabe-Thiele diagram for absorption is illustrated in Figure 8.3

Figure 8.3: A McCabe-Thiele diagram for the absorption operation

(from Separation Process Engineering by Wankat, 2007)

Note that the operating line in the McCabeThiele diagram for absorption is above the equilibrium line
18

This is because the solute is transferred from the gas phase (i.e. the gas mixture) to the liquid
phase (i.e. the solvent), which is in the opposite direction to the distillation operation, in which

the material (or the more volatile component: MVC) is transferred from the liquid phase to the gas phase

The dotted lines in Figure 8.3 is the minimum

L L line or the line; it is the line drawing G G min

from the initial point of (Xo , Y1 ) until touching

the equilibrium line at the point where Y = YN +1 (note that YN +1 is the concentration of the solute in the inlet gas stream)

19

It is important to note that, if the system is

NOT isothermal (as in the case of the distillation

operation), the operating line is still linear, but the equilibrium line is no longer linear

The following Example illustrates how to calculate the required number of equilibrium stages for the absorption operation

20

Example A gas stream containing 90 mol% N2

and 10% CO2 is passed through an absorber, in which pure and cool water at 5 oC is used as a solvent The operation is assumed to be isothermal at 5 oC and isobaric at 10 atm The liquid flow rate is 1.5 times the minimum

L liquid flow rate G

min

Determine the number of equilibrium stages required to absorb 92 mol% of CO2 Given Henrys constant of CO2 in water at 5 oC of 876 atm/mole fraction

21

Basis: 1.0 mol/h of the gas mixture

The schematic diagram for this absorption operation can be illustrated as follows

(from Separation Process Engineering by Wankat, 2007)

Substituting given numerical values into Eq. 8.4:

y= H x Ptotal
22

(8.4)

gives
876 y= x = 87.6x 10 (8.11)

Eq. 8.11 is the equilibrium equation; thus, the equilibrium or the y-x data can be computed and summarised in the following Table:
x 0 0.0001 0.0004 0.0006 0.0008 0.0010 0.0012 y 0 0.00876 0.0350 0.0526 0.0701 0.0876 0.1051

23

However, both x and y have to be converted to X and Y as exemplified below

For x = 0.0001:

x 0.0001 X= = 0.0001 1-x 1 - 0.0001 For y = 0.00876: Y = y 0.00876 = 0.00884 1 - y 1 - 0.00876

Thus, the equilibrium data on Y-X co-ordinate are as follows

X 0 0.0001 0.0004 0.0006 0.0008 0.0010 0.0012
24

Y 0 0.00884 0.0363 0.0555 0.0754 0.0960 0.1175

From the problem statement, it is required that 92 mol% of CO2 is absorbed by the solvent (i.e. cool water) Accordingly, only 8% of CO2 input into the absorber (i.e. yN +1 = 0.10 ) remains in the gas mixture, which is equal to the mole fraction of CO2 of 8 0.10 = 0.008 100 Since the basis of calculation is 1.0 mol/h of the gas mixture, the amount in mole of CO2 in the input stream is

0.10 1.0 = 0.10 mol/h

and in the output stream is

0.008 1.0 = 0.008 mol/h

25

The amount of carrier gas (i.e. N2 in this Example) is given as 90%, meaning that the flow
rate of the carrier gas (only) is

90 (1 mol/h of the gas mixture) 100 = 0.90 mole/h of carrier gas

Thus, the inlet concentration of CO2 for the gas phase (YN +1 ) is
YN +1 moles of CO 2 = moles of N
2

inlet

0.10 = = 0.11 0.90

and the outlet concentration of the CO2 in the gas phase (Y1 ) is
moles of CO 2 Y1 = moles of N
2

outlet

0.008 = = 0.0089 0.90

26

Since pure water is used as the solvent, it results in the fact that x o = 0 , which means that
Xo = 0

Accordingly, the origin of the operating line is the point Xo , Y1 of 0, 0.0089

) (

The equilibrium line (from the equilibrium data on Page 24) can be plotted on the Y-X coordinate as shown on the next Page (as a solid line)
L The G

line is the line originates from the

min

point of 0, 0.0089 and touches the equilibrium line at Y = YN +1 = 0.11; note that the X value at the touching point is 0.00105
27

0.1400 0.1200
YN+1 = 0.11

X = 0.0007
actual operating line (L/G)min line

X = 0.00105

0.1000 0.0800

Y
0.0600 0.0400 0.0200 (0, 0.0089) 0.0000

equilibrium line

0.0000

0.0002

0.0004

0.0006

0.0008

0.0010

0.0012

0.0014

L The slope of the G

line (the dotted lines)

min

is found to be 97.2

Hence, the slope of the actual operating line is

L 1.5 G = 1.5 97.2 = 145.8
28

min

The origin of the actual operating line is still

L at the point 0, 0.0089 as per the G

line

min

Since Y1 must still be the same (at 0.11) as it is the requirement, the value of X N for the actual operating line (with the slope of 145.8) can be computed as follows
slope = 145.8 = Y1 -YN +1 X N - Xo (0.11 - 0.0089)

- 0) X N 0.0007
N

(X

29

Thus, the actual operating line is the line connecting between

the point Xo , Y1 of 0, 0.0089

) (

) )

and
the point X N , YN +1 of 0.0007, 0.11

) (

Step off stages from the point Xo , Y1 of

(0, 0.0089) to the point where X = X

0.0007

yields the number of equilibrium stages of ~3.8

30

8.3 Stripping Analysis

As we have learned previously, stripping is the operation that is in the opposite direction to the absorption

In the stripping operation, the mixture of gas (i.e. the solute) and liquid (i.e. the solvent) is passed through a stripping column, in which the

gaseous component is to be stripped off from the

gas-liquid mixture by a stripping gas, as illustrated in Figure 8.4

31

Figure 8.4: A stripping operation

(from Separation Process Engineering by Wankat, 2007)

The values of Xo (the inlet concentration of the gas-liquid mixture) and YN +1 (the initial concentration of the stripping gas) are normally given (i.e. known), the value of X N (the final concentration of the treated liquid stream) is generally L is also given (either specified, and the value of G directly or indirectly)
32

Hence, we need to calculate/determine the value of Y1 (the outlet concentration of the stripping gas)

The operating line for stripping is still the

same as per the absorption; i.e.

Yj +1 L L = X j + Y1 - Xo G G (8.10)

It should be noted, however, that the stripping operation is usually NOT isothermal; hence, its

equilibrium line is normally NOT linear

The example of the McCabe-Thiele diagram for the stripping operation is as shown in Figure 8.5
33

Figure 8.5: A McCabe-Thiele diagram for the stripping operation

(from Separation Process Engineering by Wankat, 2007)

Note that the point X N , YN +1 and the point

(X , Y ) must be on the operating line

o 1

As mentioned earlier, the value of Xo is normally given, as it is the inlet concentration of the feed (the gas-liquid mixture)
34

To step off stages, we start from the intersec-

tion of the operating line and the X = Xo line

Note that, for stripping, the equilibrium curve is above the operating line; this is because the

material (i.e. a gas component in the gas-liquid mixture) migrates from the liquid phase to the
gas phase [which is the same as the more volatile component (MVC) moves from the liquid phase to the gas phase in the distillation)

Accordingly, for the stripping operation, the

maximum amount of stripping gas or the maximum L can be determined

G

35

This can be done by drawing a straight line from the point of (X N , YN +1 ) until it touches the equilibrium curve at the point where X = Xo However, it is important to note that this straight line (or the maximum L line) cannot be
G

over (or higher than) the equilibrium line; thus, sometimes, the maximum L line is the tangential
G

(but still straight) line to the equilibrium line that originates from the point (X N , YN +1 ) to the point where X = Xo , as illustrated as the dashed lines in Figure 8.6

36

8.4 Analytical Solution: Kremser Equation

When the concentration of a solute in both gas and liquid phases is very low (< 1%), the total gas and liquid flow rates do not changes

significantly, even though the solute is transferred

from the gas phase to the liquid phase

Thus, the mole fractions of species i in both gas (yi ) and liquid (x i ) phases can be used for the

calculations (in other words, it is NOT necessary

to convert that y and x data to the Y and X coordinate)

37

For the dilute situation described above, Figure 8.2 (on Page 13), which is on X and Y basis can be replaced by Figure 8.6

Figure 8.6: A gas absorption when the concentration of a solute in both gas and liquid phases are low
(from Separation Process Engineering by Wankat, 2007)

38

For a dilute absorber, the operating line is similar to that of the normal absorption operation (or Eq. 8.10), except that
Y is replaced by y (i.e. the mole fraction

of a solute in the gas phase)

X is replaced by x (i.e. the mole fraction

of a solute in the liquid phase)

G is replaced by V

Thus, the operating line for the dilute absorption can be written as follows y j +1 L L y1 - x o = xj + V V (8.12)

39

All of the assumptions are still the same as per the normal absorption (see Page 7), with an addi-

tional assumption that the concentration of the solute in both gas and liquid phases is very low
To enable the stage-by-stage problem to be solved analytically, an additional assumption must be made, and it is the assumption that the

equilibrium line is linear; i.e.

y j = mx j + b (8.13)

Actually, the Henrys law equation (Eq. 8.4): yj = HB Ptotal xj

already satisfies this assumption, especially with the assumption that this system is isobaric
40

By comparing the Henrys law equation with Eq. 8.13, it results in the fact that
m=

HB Ptotal

b=0

An analytical solution for the absorption operation can be derived for 2 special cases:
When the operating and equilibrium lines

are parallel to each other (i.e. illustrated in Figure 8.7

L = m ), as V

When the operating and equilibrium lines

are NOT parallel to each other (i.e. L < m ), as shown in Figure 8.8 V

41

Figure 8.7: The absorption operation for the case that the operating and equilibrium lines are parallel to each other
(from Separation Process Engineering by Wankat, 2007)

For the case that the operating and equilibrium lines are parallel to each other or when L = m , we obtain the fact, for the absorber with V N equilibrium stages, that yN +1 - y1 = N Dy where
42

(8.14)

Figure 8.8: The absorption operation for the case that the operating and equilibrium lines are not parallel to each other
(from Separation Process Engineering by Wankat, 2007)

Dy = y j +1 - y j
in which

(8.15)

y j +1 is obtained from the operating line

(Eq. 8.12)
y j is obtained from the equilibrium line

(Eq. 8.13)
43

Combining Eqs. 8.12 & 8.13 with Eq. 8.15 and re-arranging yields

L L Dy = - m + x y x b 1 V o j V
(8.16)

L Since, in this case, = m , Eq. 8.16 becomes V L Dy = y1 - xo - b = constant V (8.17)

Note that Eq. 8.17 is true because the opera-

ting and equilibrium lines are parallel to each

other; thus, the distance between the operating and the equilibrium lines or Dy are constant

44

Substituting Eq. 8.17:

L Dy = y1 - xo - b V
into Eq. 8.14:

(8.17)

yN +1 - y1 = N Dy
L = yN +1 - y1 N y1 - xo - b V

(8.14)

and re-arranging the resulting equation yields

N =

L y1 - x o - b V
(8.18)

yN +1 - y1

Eq. 8.18 is a special case of the Kremser equa-

L tion, when =m V

L = 1 or mV
45

L For the case where < m (see Figure 8.8 on V Page 43), Dy is no longer constant or Dy varies from stage to stage

Re-arranging Eq. 8.13:

y j = mx j + b
results in
xj = yj -b m

(8.13)

(8.19)

Substituting Eq. 8.19 into Eq. 8.16:

L L x y x b Dy = - m + j 1 o V V
(8.16) and re-arranging for Dy of the stages j and j + 1 yields
46

(Dy )
and

L L L = - 1 y j + y1 b - xo mV V mV
(8.20)

(Dy )

j +1

L L L = - 1 + y y1 b - xo j +1 mV V mV
(8.21)

(8.21) (8.20) and re-arranging gives

(Dy )

j +1

L = Dy ) ( j mV

(8.22)

Eq. 8.22 provides the relationship of Dy for the adjacent stages (stage ; e.g., stage 1

L and stage 2) with the coefficient of (commV monly called the absorption factor)
47

Since Dy is NOT constant, Eq. 8.14 has to be modified to

yN +1 - y1 = Dy1 + Dy2 + Dy 3 ... + DyN

(8.23) By employing Eq. 8.22, we obtain the fact that, e.g.,

(Dy )

(Dy )

L = Dy ) ( 1 mV L = Dy ) ( 2 mV

(8.24) (8.25)

Substituting Eq. 8.24 into Eq. 8.25 yields

(Dy )

L = mV

L Dy ) ( 1 mV
2

(Dy )

L = Dy ) ( 1 mV
48

(8.26)

Eq. 8.26 can be written in a general form as

(Dy )

j +1

L = Dy ) ( 1 mV

(8.27)

By applying Eq. 8.27 with Eq. 8.23:

yN +1 - y1 = Dy1 + Dy2 + Dy 3 ... + DyN

(8.23) we obtain the following equation:
2 N -1 L L L + ... + yN +1 - y1 = 1 + + Dy1 mV mV mV

(8.28)

After performing mathematical manipulation for the summation on the right hand side (RHS) of Eq. 8.28 and re-arranging the resulting equation, it results in

49

 L 1- yN +1 - y1 mV = L Dy1 1- mV

(8.29)

Writing Eq. 8.22:

(Dy )
for stage 1 gives

j +1

L Dy ) ( j mV

(8.22)

(Dy )

L = Dy ) ( o mV

(8.30)

* (see Figwhere (Dy ) = Dy (at x = xo ) = y1 - y1 o

ure 8.8 on Page 39)

Note that, from the equilibrium line/equation,

* y1 = mx o + b

(8.31)

50

Thus, Eq. 8.30 can be re-written as

(Dy )

L * y1 - y1 = mV

(8.32)

Substituting Eq. 8.32 into Eq. 8.29 and rearranging gives

yN +1 - y1 L * y1 - y1 mV L 1- mV = L 1- mV
N

L L - yN +1 - y1 mV mV = * L y1 - y1 1- mV

N +1

(8.33)

51

Solving Eq. 8.33 for N yields

* y y mV mV 1 N +1 + ln 1 * L L y y 1 1 N = mV ln L

(8.34)

Note that, in this case,

L L m or 1 V mV

Eq. 8.34 is another form of the Kremser equation, when

L 1 mV

The Kremser equations in terms of gas-phase composition can be written as follows

52

yN +1 - y1 y1 - y
* 1

L L mV mV L 1- mV
L = mV
N

N +1

(8.33)

yN +1 - y y1 - y

* N +1 * 1

(8.35)

* y y N +1 N +1 ln * y1 - y1 N = L ln mV * y y N +1 N +1 ln * y1 - y1 N = y y ( N +1 1) ln * * yN +1 - y1

(8.36)

(8.37)

* * = mx + b and y = mx o + b where yN +1 1 N

53

The Kremser equations in terms of liquidphase composition are

* x x L L o N + ln 1 * mV mV x x N N N = mV ln L

(8.38)
x - x * N N ln * x x o o N = x * - x * o N ln x x o N mV 1 * xN - xN L = N +1 * xo - x N mV 1- L

(8.39)

(8.40)

54

xN - x xo - x
* N

* N * o

L = mV

(8.41)
* o

where x =

yN +1 - b m

and x =

y1 - b m

Among many equations above, how to choose the appropriate equation(s) depends on the data given in the problem statement

Lets examine the following Example, which illustrates how to select the proper Kremser equation for a specified problem

55

Example A plate tower with 6 equilibrium stages

is employed for stripping ammonia from waste water with the inlet concentration of 0.10 mol% using counter-current air at atmospheric pressure and 80 oF Determine the concentration of ammonia in the exit waste water if the stripping air is ammo-

nia-free and the feed rate of air is 30 standard

cubic feet (scf) per 1 lbm of waste water Given the equilibrium equation of ammonia at 80 oF as y = 1.414x

56

One lb-mol of air is equivalent to the volume of 379 standard cubic feet (scf) [for 60 oF (519.67 R) and 14.7 psi (lbf/in2)], which can be computed as follows
nRT V = P

3 psia ft ( ) 1 lb-mol 10.73 ( ) (60 + 459.67 ) R lb-mol)(R ) ( = (14.7 psia )

( )

V = 379 ft3

The molecular weight (MW) of water is 18.02

lbm lb-mol

57

Thus, the ratio of the volumetric flow rate of the stripping gas or air (V ) to that of the waste
water (L ) can be calculated as follows
30 scf air 1 lb-mol air lb water V m = 18.02 L lb-mol water 1 lbm water 379 scf air

lbm - mol air V = 1.43 lb-mol water L The given data are (see the following Figure)

(from Separation Process Engineering by Wankat, 2007)

58

 The inlet concentration of ammonia (NH3)

0.1 in waste water: x o = = 0.001 100 The inlet concentration of NH3 in the stripping gas or air: yN +1 = y7 = 0 (as the air is ammonia-free)
m = 1.414 and b = 0 [from the given

equilibrium equation: y = 1.414x , compared to the standard equilibrium equation (Eq. 8.13): y = mx + b ]
The number of equilibrium stages (N ) = 6

Thus, by employing Eq. 8.40:

* xN - xN

xo - x

* N

mV 1- L mV 1- L
59

N +1

(8.40)

with the fact that

x =
* N

yN +1 - b m

which yields 0-0 x = =0 1.414 we obtain the value of the outlet concentration
* N

of NH3 in the waste water (x N or x 6 ) as follows

mV 1 * * L xN - xN x x = o N N +1 mV 1 - L 1 - (1.414)(1.43) (0.0010 - 0) xN - 0 = + 6 1 1 - (1.414)(1.43)

x N = 7.45 10-6 = 7.45 ppm

60