Process Separations

ChE 3131

Chapter 6 - Absorption and Stripping

Materials are taken from the Textbook

SEPARATION PROCESS PRINCIPLES with Applications Using
Process Simulators, Seader, J. D., Henley, Ernest J., Roper, D.
Keith, 4th Edition, John Wiley & Sons, Inc., 2016.

Slides are Prepared by Professor Muthanna Al-Dahhan

General Introduction
• Absorption uses a liquid to
(1) Selectively remove components from a gas mixture;
(2) Remove impurities, contaminants, pollutants, or catalyst poisons; and
(3) Recover valuable chemicals.
• The species of interest in the gas mixture may include all components present in the gas, only the component(s) not
absorbed, or only the component(s) absorbed.
• The species transferred to the liquid absorbent are called solutes. When the solutes have been absorbed, they are called
absorbates.
• In stripping (desorption), a liquid mixture is brought into contact with a gas that selectively removes components by mass
transfer from the liquid to the gas phase.
• Strippers are frequently coupled with absorbers to permit the recovery and recycling of the absorbent.
• When water is used as the absorbent, it is common to separate the water from the solute by distillation rather than by
stripping.
Industrial Example – Figure 6.1 Scrubbed Gas
Almost saturated with water vapor
• The feed gas is air containing water vapor and acetone vapor.
• It comes from a dryer in which water and acetone are evaporated from solid
cellulose acetate fibers.
• The 30-tray (equivalent to 12.5 equilibrium stages) countercurrent absorber is
designed to remove 99.5% of the acetone from the feed gas. 148 ppm
• Acetone is the main species absorbed, along with minor amounts of O2 and N2.
• Water present as vapor in the feed gas and as liquid in the entering absorbent is
absorbed and stripped, respectively.
• The net effect is that water is stripped because more water appears in the exit
gas than in the feed gas.
• The amount of each component absorbed depends on the number of equilibrium
stages and the component’s absorption factor, Almost saturated with air

• When water or hydrocarbon oils are used as absorbents and no significant chemical reactions occur between the absorbent
and the solute, the process is called physical absorption
• When aqueous NaOH is used as the absorbent for an acid gas, absorption is accompanied by a rapid and irreversible reaction
in the liquid. This is chemical absorption or reactive absorption.
• More complex examples are the processes for absorbing CO2 and H2S with aqueous solutions of monoethanolamine (MEA)
and diethanolamine (DEA), during which there is a more desirable reversible chemical reaction in the liquid.
• Chemical reactions can increase the rate of absorption and solvent-absorption capacity and convert a hazardous chemical to
an inert compound
• For absorption and stripping, design procedures are well developed and commercial processes are common. Table 6.1 lists
representative applications. Environmental standards for the removal of pollutants have greatly increased the use of gas
absorbers.

• Equilibrium-based and rate-based (mass-transfer) models, using both graphical and algebraic procedures, for physical
absorption and stripping of mainly dilute mixtures are described.
• The methods also apply to reactive absorption with irreversible and complete chemical reactions of solutes with solvents.
• Calculations for concentrated mixtures and reactive absorption with reversible chemical reactions are best made with
process simulators, as discussed in Chapters 10 and 12.
6.1 EQUIPMENT FOR VAPOR–LIQUID SEPARATIONS
• Methods for designing and analyzing absorption, stripping, and distillation depend on the type of equipment used for
contacting vapor and liquid phases.
• When multiple stages are required, phase contacting is mostly carried out in equipment of the type shown in Figure 6.2.
• The most common devices are cylindrical, vertical columns containing trays or packing.
• Less common are spray towers, bubble columns, and centrifugal contactors.

(a) trayed tower (c) spray tower (d) bubble column (e) centrifugal contactor
(b) packed column

Figure 6.2 Industrial equipment for absorption and stripping: (a) trayed tower; (b) packed column; (c) spray tower; (d) bubble
column; (e) centrifugal contactor
6.1.1 Trayed Columns
• A trayed tower is a vertical, cylindrical pressure vessel in which vapor and liquid, flowing countercurrently, are contacted on trays (plates)
that provide intimate contact of liquid with vapor in a froth to promote rapid mass transfer (Figure 6.3).
• Phase disengagement occurs above the froth and below the tray above.
• Liquid flows across each tray, over an outlet weir, and into a downcomer, which takes the liquid by gravity to the tray below.
• Gas flows upward through openings in each tray, bubbling through the liquid on the tray to produce a froth.
• When vapor flows through the holes in the tray, any of five two-phase-flow regimes shown in Figure 6.4 may occur alone, or in combination
• In the spray regime, the gas phase is continuous. Jets of vapor rise from the tray openings through the liquid on the tray, entraining and
spraying some of the liquid into the space between trays.
• This regime occurs for low weir heights (low liquid
depths) at high gas rates.
• The most common and favored regime is the froth
regime, in which both liquid-continuous and gas-
continuous dispersions can coexist.
• The gas passes through the froth in the form of
jets or a series of bubbles.
• At high liquid rates, small gas bubbles may be
undesirably emulsified. If bubble coalescence is
hindered, undesirable foam forms.
• Ideally, the liquid carries no vapor bubbles
(occlusion) to the tray below, the vapor carries no
liquid droplets (entrainment) to the tray above,
and there is no weeping of liquid through the
holes in the tray.
• With good contacting, equilibrium between the
exiting vapor and liquid phases is approached on
each tray, unless the liquid is very viscous.
• Figure 6.5 shows typical tray openings for vapor passage: (a) perforations, (b)
valves, and (c) bubble caps.
• The simplest is the sieve or perforated tray with holes, usually 1/8 to 1/2-inch in
diameter.
• A valve tray has openings commonly from 1 to 2 inches in diameter. Each hole is
fitted with a valve consisting of a cap that overlaps the hole, with legs or a cage
to limit vertical rise, while maintaining the valve cap in a horizontal orientation.
Without vapor flow, each valve covers the hole. As the vapor rate increases, the
(a) perforation (b) valve cap (c) bubble cap
valve rises, providing a larger and larger opening through which the vapor can
flow to create a froth.
• A bubble-cap tray consists of a cap, 3 to 6 inches in diameter, mounted over and
above a concentric riser, 2 to 3 inches in diameter. The cap has rectangular or
triangular slots cut around its periphery.
• The vapor flows up through the tray opening into the riser, turns around, and
passes out through the slots and into the liquid, forming a froth.
• Improvements are constantly being made to increase the efficiency and
throughput of sieve and valve trays.
• Recent changes have increased tray efficiency by several percent and capacity by d) tray with valve caps
more than 20%.

Figure 6.5 Three types of tray openings for the

passage of vapor up into the liquid: (a) perforation;
(b) valve cap; (c) bubble cap; (d) tray with valve
caps.
6.1.2 Packed Columns
• A packed column, is a vertical, cylindrical vessel containing one or more sections of
packing over whose surface the liquid flows downward as a film on the packing and
walls, or as droplets between packing elements.
• Feed gas enters at the bottom, passes through a vapor distributor, and flows upward
through the wetted packing, thus contacting the liquid and passing out the top.
• Liquid enters at the top into a liquid distributor, flows downward through the packed
sections, and leaves at the bottom
• Sections containing random packings and structured
packings
• A liquid redistributor ensures uniform distribution of liquid
over the cross-sectional area of the column as it enters a
packed section
• liquid may channel downward near the wall, where
resistance is the lowest, while gas flows up the center of the
column. Channeling significantly reduces vapor–liquid
contact and increases the height equivalent to a theoretical
plate (HETP)
• Commercial packing materials include
(1) Random (dumped) packings
(2) Structured packings of crimped layers of mesh or
corrugated sheets

Figure 6.8 Typical structured

packings of the last three of four
generations. Reproduced with Figure 6.6 Details of internals in a column
permission of Sulzer Chemtech Ltd. Figure 6.7 Four generations of random packings with packing and trays
6.1.3 Choice between Trays and Packing
• For contacting vapor and liquid streams for absorption, stripping, and distillation, the designer must choose between a trayed column and a
packed column.
• Random packing is favored when the column diameter is less than 2 ft and the packed height is less than 20 ft.
• Packed columns are favored for
(1) Corrosive services that tend to use ceramic or plastic materials rather than metals, particularly welded column internals;
(2) Services that produce foaming too severe for the use of trays;
(3) Vacuum operations and other services for which pressure drop must be low; and
(4) Services that benefit from low liquid holdup in the column. Otherwise, trayed towers, which can be designed more reliably, are
preferred.
• Although structured packings are expensive, they are the best choice for installations when pressure drop is a factor or for replacing existing
trays (retrofitting) when a higher capacity or degree of separation is required.
• Trayed towers are preferred when liquid velocities are low, whereas columns with random packings are best for high liquid velocities.
• Use of structured packing should be avoided at pressures above 200 psi and liquid flow rates above 10 gpm∕ft2 (Kister [33]).
• Turbulent liquid flow is desirable if mass transfer is limiting in the liquid phase, and a continuous, turbulent gas flow is desirable if mass
transfer is limiting in the gas phase.
• Usually, the (continuous) gas phase is mass-transfer-limiting in packed columns and the (continuous) liquid phase is mass-transfer-limiting in
tray columns.
6.2 GENERAL DESIGN CONSIDERATIONS
• Absorber and stripper design or analysis requires consideration of the following factors:
1. Entering feed gas (or liquid) flow rate, composition, T, and P
2. Desired degree of recovery of one or more solutes
3. Choice of absorbent (or stripping agent)
4. Column operating P and T, and allowable gas pressure drop
5. Minimum absorbent (or stripping agent) flow rate and actual absorbent (or stripping agent) flow rate
6. Heat effects and need for cooling (or heating)
7. Number of equilibrium stages and stage efficiency
8. Type of absorber (or stripper) equipment (trays or packing)
9. Need for liquid redistributors if packing is used
10. Height of absorber (or stripper)
11. Diameter of absorber (or stripper)
• The ideal absorbent has
(a) a high solubility for solute(s); • Absorber operating pressure should be high and temperature low in
(b) a low volatility to reduce loss; order to minimize stage requirements and/or absorbent flow rate, and
(c) stability and inertness; to lower the equipment volume required to accommodate the gas flow
(d) low corrosiveness; • Operating pressure should be low and temperature high for a stripper
(e) low viscosity and high diffusivity; to minimize stage requirements and stripping agent flow rate
(f) low foaming proclivities;
(g) low toxicity and flammability;
(h) availability, if possible, within the process; and
(i) low cost.
• The most widely used absorbents are water, hydrocarbon oils, and aqueous solutions of acids and bases. The most common stripping
agents are steam, air, inert gases, and hydrocarbon gases.
6.3 GRAPHICAL METHOD FOR TRAYED TOWERS
• Stages are conventionally numbered from the top, where the absorbent
enters, to the bottom.
• For the stripper, stages are numbered from the bottom, where the stripping
agent enters, to the top
• Phase equilibrium is assumed between the vapor and liquid leaving each tray.
• Assume for an absorber that only one solute transfers.
Let:
L′ = molar flow rate of solute-free absorbent
V′ = molar flow rate of solute-free gas (carrier gas)
X = mole ratio of solute to solute-free absorbent in the liquid
Y = mole ratio of solute to solute-free gas in the vapor
• With these definitions, the values of L′ and V′ remain constant throughout
the tower because only solute undergoes mass transfer between phases
6.3.1 Equilibrium Curves for the Solute
• For the solute at any stage n, the K-value is

where Y = y∕(1 − y) and X = x∕(1 − x)

• The equilibrium curve will not be a straight line, but it will pass
through the origin.
6.3.2 Operating Lines (from Solute Material Balances)
• Solute material balances are written around one end of the tower and an arbitrary
intermediate equilibrium stage, n.
• For the absorber, the total solute flow rate into the envelope equals the total solute
flow rate out of the envelope in Figure 6.9a:
n+1

operating line

operating line n+1

• Equations (6-3) and (6-5) are the operating lines plotted in Figure 6.9.
• The terminal points represent conditions at the top and bottom of the tower.
• For absorbers, the operating line is above the equilibrium line because, for a given
solute concentration, X, in the liquid, the solute concentration, Y, in the gas is always
greater than the equilibrium value, thus providing a mass-transfer driving force for
absorption.
• For strippers, operating lines lie below equilibrium lines, thus enabling desorption.
• Operating lines are straight with a slope of L′ ∕ V′ .
6.3.3 Minimum Absorbent Flow Rate (for ∞ Stages)
• Operating lines for four different solute-free absorbent flow rates, L′ , are shown in
Figure 6.10 for a fixed solute-free gas feed rate, V′ . In each case, the solute
concentration in the exiting gas, Y1, is the same.
• Therefore, each operating line passes through the terminal point, (Y1, X0), at the top of
the column.
• To achieve the desired value of Y1 for given YN+1, X0, and V′ , the solute-free
absorbent flow rate L′ must be between an ∞ absorbent flow with L′ ∕ V′= ∞, as
represented by operating line 1, and a minimum absorbent rate (corresponding to ∞
stages), L′ min, as represented by operating line 4, with the equilibrium curve and
operating line intersecting at YN+1.
• Intermediate operating lines 2 and 3, correspond to 2 and 1.5 times L′ min, respectively.
The solute concentration in the outlet liquid, XN, depends on L′ .
• Using a solute material balance over the entire absorber, (6-2) with n = N gives

• Note that the operating line can terminate at the equilibrium line as in operating
line 4, but cannot cross it because that would be a violation of the second law of
thermodynamics.
• The minimum absorbent flow rate, L′ min, corresponds to a value of XN (leaving the
bottom of the tower) in equilibrium with YN+1, the solute concentration in the feed gas.
Note that it takes an infinite number of stages for this equilibrium to be achieved. An
expression for L′ min of an absorber can be derived from (6-7) as follows. For stage N, (6-1)
for the minimum absorbent rate becomes
Pinch
• Equation (6-11) confirms that Point
L′min increases with increasing
V′, K, and fraction of solute
absorbed.
• In practice, the absorbent flow
rate is some multiple of L′ min,
typically from 1.1 to 2.
• As the operating line moves from
1 to 4, the number of stages
required increases from zero to
infinity. Thus, a trade-off exists
between L′ and N, and an
optimal L′ exists.

• A similar derivation of V′ min, for

Operating Line Equilibrium Line
the stripper of Figure 6.9, is
analogous to (6-11):

• In practice, stripping factors, defined by S = KV∕L, are optimal at about 1.4.

6.3.4 Number of Equilibrium Stages
• As shown in Figure 6.11a, the operating line
relates the solute concentration in the vapor
passing upward between two stages to the
solute concentration in the liquid passing
downward between the same two stages.
• Figure 6.11b illustrates that the equilibrium
curve relates the solute concentration in the
vapor leaving an equilibrium stage to the solute
concentration in the liquid leaving the same
stage.
• This suggests starting at the top of the absorber
(at the bottom of the Y–X diagram) and moving
to the bottom of the absorber (at the top of
the Y–X diagram) by constructing a staircase
that alternates between the operating line and
the equilibrium curve, as in Figure 6.12a

(Y2, X1)
Operating Line
Equilibrium Line
2 2
(Y2, X2)
Absorber
• Moving up the staircase, steps off the number of stages required for a given solute-free absorbent flow rate corresponding to the slope of
the operating line, which in Figure 6.12a is
(L′ ∕V′ ) = 1.5 (L′min∕V′ )
• Starting at the point (Y1, X0) on the operating line, a horizontal move is made to the right to (Y1, X1) on the equilibrium curve.
• From there, a vertical move is made to (Y2, X1) on the operating line.
• The staircase is climbed until the terminal point (YN+1, XN) on the operating line is reached.
• This point corresponds to the passing streams at the bottom end of the column.
• As shown in Figure 6.12a, the stages are counted at (YN+1, XN)
the points on the equilibrium curve. In this case,
N = 3.
• As the operating-line slope L′∕ V′ is increased, fewer
equilibrium stages are required. (Y3, X2)
• As L′∕ V′ is decreased, more stages are required
until L′min ∕ V′ is reached, at which the operating 2
3
line and the equilibrium curve intersect at a pinch
(Y2, X1) (Y2, X2)
point located on the horizontal line for YN+1.
• At L′min, an infinite number of stages is required.
Operating line 4 in Figure 6.10 shows a pinch point
at (YN+1, XN). (Y1, X1)
• If L′∕ V′ is reduced below L′min ∕ V′ the specified
extent of absorption cannot be achieved.

a Absorber

Figure 6.12 Graphical determination of the number of

equilibrium stages for (a) absorber
Stripper
• The stages required for stripping a solute are determined analogously to absorption.
• An illustration is shown in Figure 6.12b.
• For given specifications of Y0, XN+1, and the extent of stripping of the solute, X1, V′ min is determined from the slope of the operating line
that passes through the points (Y0, X1) and (YN, XN+1)
• The operating line in Figure 6.12b is for: V′ = 1.5 V′min
or a slope of (YN, XN)
(L′ ∕ V′ )=(L′ ∕ V′min) ∕ 1.5 (Y3, X3)
(YN, XN+1)
• In Figure 6.12, the number of equilibrium stages for
both the absorber and stripper is exactly 3.
• Ordinarily, the result is some fraction above an integer
(Y2, X2)
number, and the result is usually rounded to the next
highest integer to ensure achievement of the desired (Y2, X3) 3
composition.
2
(Y1, X1)
(Y1, X2)

(Y0, X1)

b Stripper
Figure 6.12 Graphical determination of the number of
equilibrium stages for (b) stripper.
6.4 KREMSER GROUP METHOD FOR MULTICOMPONENT ABSORPTION AND STRIPPING
• The graphical method of section 6.3 for determining N has educational value because a graph provides a visual insight into the phenomena
involved.
• However, the graphical method becomes tedious when the following occurs:
(1) Problem specifications fix the number of stages rather than the percent recovery of solute,
(2) More than one solute is absorbed or stripped,
(3) The location of the equilibrium curve is unknown because the optimal operating T and P have not been established, and
(4) Very low or very high concentrations of solute force the construction to the corners of the diagram so that multiple Y–X diagrams of
varying scales are needed to achieve accuracy.
• When the graphical method is unsuitable, the analytical method of Kremser is useful for obtaining results that can be used to initialize
rigorous methods in process simulators.
• Kremser [30] introduced a group method that relates the number of equilibrium stages to the recovery of one key component in a single-
section, countercurrent cascade used for multicomponent absorption or stripping.
• The procedure is called a group method because it only provides an overall treatment of the group of stages in the cascade.
• The procedure does not consider detailed changes in temperature, pressure, phase compositions, and flow rates from stage to stage.
• The treatment here is similar to that of Edmister [31].
• Consider a countercurrent absorber of N adiabatic, equilibrium stages, as shown in
Figure 6.14a, with stages numbered from top to bottom. L0=0 v1
• The absorbent is pure, and component molar flow rates are vi and li. In the vapor
and liquid phases, respectively.
• Mole fractions are y and x, and total molar flow rates are V and L.

• A group method is to do balance on a group of stages such as between stage 1 and N-1,
then stage 1 and N-2, then stage 1 and stage N-3 and so on. A parameters will be derived vN lN+1 N
such as effective absorption factor, Ae, absorption recovery factor, ϕA, (i.e., fraction of
species not absorbed, fraction of species not stripped, stripping factor)
Figure 6.14 Countercurrent cascades of N adiabatic stages: (a) absorber
• The following derivation applies to any component in the vapor feed.
• A material balance around the top, including stages 1 through N − 1, for any
absorbed species is as follows, with the subscript for species, i, dropped for
convenience:

L0=0 v1

Figure 6.14
Countercurrent
cascades of N adiabatic
stages: (a) absorber
vN N
N lN+1
• The component flow rate, lN-1, is eliminated by successive substitution using material balances around successively smaller sections of the
top of the cascade. For stages 1 through N−2,
• In the Kremser method, an average, effective absorption factor, Ae, for each
component replaces the separate absorption factors for each stage,
simplifying (6-27) into

• Because each component has a different Ae, ϕA is also component

specific.
• Figure 6.15 from Edmister [31] is a plot of (6-29) with a probability scale
for ϕA, a logarithmic scale for Ae, and N as a parameter.
• By specifying any two of these three quantities, the third can be
determined from the plot.
• Kremser first developed this plot, using linear coordinates [30].
• The plot of the Kremser equation in Figure 6.15 shows a strong
dependence for the effect of the number of equilibrium stages, N,
(referred to as theoretical plates in Figure 6.15) on the fraction not
absorbed, ϕA, for a given absorption factor, A.
• Components with high absorption factors (i.e., greater than the
optimal value of 1.4) show a rapidly decreasing fraction not
absorbed for an increasing number of stages.
• These are components that are easily absorbed, as indicated by their
low K-values.
• As the absorption factor decreases below 1.4, the effect of an
increase in the number of stages diminishes rapidly.
• Below an absorption factor of 0.5, there is little effect on the fraction
not absorbed above 3 or 4 stages.
• These are components that are difficult to absorb, as evidenced by
their high K-values.
• In this region, the maximum value of the fraction that can be
absorbed is equal to the absorption factor. That is, (1-ϕA)max = A.
• Because the Kremser plot for absorption is also applicable for
stripping, the above analysis applies equally to strippers.
• This plot of an approximate analytical solution relates the state of a
system, K = K{T, P, xi, yi}, and the operability (L, V) and geometry (N)
of corresponding equipment to performance (ϕ) in a way that is
intuitive and useful to guide design and operation of a separation
system.
Stripper
• The stripper is shown in Figure 6.14b.
• Assume the components stripped from the liquid are not present in the entering vapor, and ignore absorption of the stripping agent.
• Stages are numbered from bottom to top.
• The pertinent stripping equations are derived in a manner analogous to the absorber equations. The results are

Figure 6.14 Countercurrent

cascades of N adiabatic stages:
(a) absorber; (b) stripper
• Figure 6.15 also applies to Equation (6-31)
• Absorbers are frequently coupled with strippers or distillation columns to permit the
recovery and recycling of absorbent.
• Because the stripping action is not perfect, recycled absorbent contains species present
in the vapor feed.
• Up-flowing vapor strips these species, as well as others in the makeup absorbent.

Figure 6.16 Various coupling schemes for

absorbent recovery: (a) use of steam or inert gas
stripper; (b) use of reboiled stripper; (c) use of
distillation
• A more general absorber equation for handling this situation is obtained by combining (6-26) for absorption with a form of (6-30) for
stripping species from the entering liquid.
• For a species that appears only in the entering liquid of an absorber, with stages numbered from top to bottom, as in Figure 6.14a, (6-30)
becomes

• Finally, for a component appearing in both entering vapor and entering liquid, a material balance is obtained by adding (6-26) and (6-34),
giving

• which applies to each component in the entering vapor. The analogous equation to (6-35) for a component appearing in both entering vapor
and entering liquid of a stripper is

• In order to use (6-29) and (6-31) with Figure 6.15, effective values of absorption and stripping factors, Ae and Se, are needed.
• For preliminary calculations prior to rigorous calculations with a process simulator (as described in Chapter 10) molar L and V flow rates may
be taken as entering values.
• K-values depend mainly on T, P, and liquid-phase composition.
• At near-ambient pressure, for dilute mixtures, some common expressions are:

Raoult’s law is for ideal solutions involving solutes at subcritical temperatures

The modified Raoult’s law is useful for nonideal solutions for which activity
coefficients are known at or near infinite dilution.
For solutes at supercritical temperatures, the use of Henry’s law may be preferable

For sparingly soluble solutes at subcritical temperatures, (6-40) is preferred when

solubility data are available
• The expression

is derived by considering a three-phase system that consists of an ideal vapor containing the solute, carrier vapor, and solvent;
a pure or near-pure solute as liquid (1); and a solvent liquid (2) with dissolved solute.
• In that case, for solute i at equilibrium between the two liquid phases,

• For moderate to high pressures, vapor fugacity coefficients must be included

The computations, made with a spreadsheet, produce the following results:
6.5 STAGE EFFICIENCY AND COLUMN HEIGHT FOR TRAYED COLUMNS
• Stage efficiency concepts are applicable when phases are contacted and then separated—that is, when discrete stages and interfaces are
identified as in trayed columns. This is not the case for packed columns which is embedded into equipment- and system-dependent
parameters,
6.5.1 Overall Stage Efficiency where Eo is the fractional overall stage efficiency, usually less than 1.0; Nt is the
calculated number of theoretical stages; and Na is the actual number of trays
required.
6.5.2 Tray Efficiencies from Column Performance Data
• Performance data from industrial absorption and stripping trayed columns generally include feed and product flow rates and compositions,
pressures and temperatures at the bottom and top of the column, details of the tray design, column diameter, tray spacing, average liquid
viscosity, and computed overall tray efficiency with respect to one or more components.
• From these data, particularly if the system is dilute with respect to the solute(s), the graphical or algebraic methods described in sections 6.3
and 6.4 (KREMSER GROUP METHOD) can estimate the number of equilibrium stages, Nt. Then, knowing Na, Equation (6-41) can be applied
to determine the overall stage efficiency, Eo.
• Values for absorbers and strippers are typically low, especially for absorption, for which Eo is often less than 50%, especially if the absorbent
is a hydrocarbon oil.
• Individual component overall efficiencies differ because of differences in component physical properties.
6.5.3 Empirical Correlations for Tray Efficiency
• O’Connell [8] correlation
• Drickamer and Bradford [6] Correlation

where Eo is in percent and μ is in cP, fits the data with an average

deviation of 10.3%. Equation (6-42) should not be used for non-
hydrocarbon liquids and is restricted to the viscosity range of the where Eo is in percent and other variables
data. have the units shown in Figure 6.19

• Edmister [10] and Lockhart et al. [11] suggested slight modifications to the O’Connell correlation, shown in Figure 6.19, to permit its use with
K-values (instead of Henry’s law constants).
6.5.4 Semitheoretical Models: Murphree Efficiencies
• The Murphree vapor efficiency, which is the oldest and most widely used, is derived with the assumptions of the vapor and liquid streams
entering each tray are of uniform composition, a uniform liquid composition on the tray equal to that leaving the tray, and a plug flow of the
vapor passing up through the liquid. It applies to trayed columns for distillation, as well as for absorption and stripping.
6.5.5 Tray Efficiencies by Scale-up of Data with the Oldershaw Column
• When vapor−liquid equilibrium data are unavailable, particularly if the system is highly nonideal with possible formation of azeotropes, tray
requirements and the feasibility of achieving the desired separation must be verified by laboratory tests using Oldershaw column which is a
useful apparatus of small glass or metal sieve-plate column with center-to-side down-comers developed by Oldershaw [22].
6.5.6 Column Height
• The number of trays, Na, and the tray spacing determine the height of a column between the top tray and bottom tray.
• The total column height is estimated by adding another 4 ft above the top tray for removal of entrained liquid and 10 ft below the bottom
tray for bottoms liquid surge capacity.
• If the height is greater than 250 ft, two or more columns arranged in series may be preferable to a single column.
6.6 FLOODING, COLUMN DIAMETER, AND TRAY LAYOUT FOR TRAYED COLUMNS
• When trays are designed properly: (1) Vapor flows only up through the open regions of the tray
outside the downcomers. (2) Liquid flows from tray to tray only through downcomers. (3)
Liquid neither weeps through the tray perforations nor is carried by the vapor as entrainment
to the tray above. (4) Vapor is neither carried (occluded) downward by the liquid in the
downcomer nor allowed to bubble up through the liquid in the downcomer.
• Tray layout design determines tray diameter and division of total tray cross-sectional area, AT ,
into bubbling area, Ab, and liquid downcomer area, Ad, both from the tray above and to the
tray below.
6.6.1 Flooding and Tray Diameter
• Four limits are shown (Figure 6.25): (1) excessive liquid entrainment at low liquid rates, (2)
entrainment (jet) flooding at high vapor rates, (3) downcomer (choke) flooding at high
liquid rates, and (4) weeping of liquid to the tray below at low vapor rates.
Tray Diameter
• For well-designed trays, column diameter is determined by entrainment flooding by the carry-up of suspended droplets by rising vapor or
the throw-up of liquid by vapor jets at tray perforations, valves, or bubble-cap slots
6.6.2 High-Performance Trays
• Since the 1990s, high-performance trays have been retrofitted and newly installed in many industrial columns. Changes to conventional
sieve and valve trays have led to capacity increases of more than 20% of that predicted for sieve and valve trays. Tray efficiencies have also
been increased somewhat.
6.6.3 Tray Layout
• When the tray diameter is known, the tray layout can be created.
• The layout consists of the selection of tray type (sieve or valve); the number of liquid passes across the tray; the tray spacing; the
downcomer area; the downcomer weir height at the top; and the downcomer clearance at the bottom.
• These parameters are determined so as to keep both weeping and entrainment below 10% of the liquid flow rate.
• Equivalent clear-liquid height in the downcomer should not exceed 50% of the tray spacing at low to moderate pressures and should be less
than 35% at high pressures.
• Tray pressure drop should be approximately 0.05 psi (0.35 kPa) for vacuum operation and no more than 0.15 psi (1.07 kPa) at high
pressures.
• Weir loading should not be greater than 8 gpm/inch of weir length.
• For sieve trays, the percent hole area and hole diameter must be selected.
• For valve trays, the type of valve and valve spacing are needed.
• Preliminary tray layout calculations are best made with a process simulator
6.7 RATE-BASED METHOD FOR PACKED COLUMNS
• Packed columns are continuous, differential-contacting devices that do not have physically distinguishable, discrete stages.
• Thus, packed columns are better described by mass transfer models than by equilibrium-stage concepts.
• However, in practice, packed-tower performance is often presented on the basis of equivalent equilibrium stages using a
packed height equivalent to a theoretical stage, the HETP or HETS, and defined by the equation

• The HETP concept has no theoretical basis.

• Accordingly, although HETP values can be related to mass-transfer coefficients, such values are best obtained by back-
calculation from (6-68), using experimental data.
• Usually, the lower the HETP, the more expensive and complex the packing.
Packed Height
• It is preferable to determine the packed height from theoretically based
methods involving mass-transfer coefficients.
• Consider the countercurrent-flow packed columns of packed height lT
shown in Figure 6.29.
• For packed absorbers and strippers, operating-line equations analogous
to those in section 6.3.2 of trayed column can be derived.
• For the absorber in Figure 6.29a, a molar material balance for the solute, around
the upper envelope of packed height, l, gives

• In (6-69) to (6-71), mole fractions y and x represent bulk compositions of

the gas and liquid in contact at any vertical location in the packing.
• For the case of absorption, with solute mass transfer from the gas to the liquid
stream, the two-film theory of Chapter 3, illustrated in Figure 6.30, applies.
• A concentration gradient exists in each thin film.
• At the interface between the two phases, physical equilibrium exists.
• Thus, as with trayed towers, an operating line and an equilibrium line are of
great importance for packed towers.
• The location of the two lines is independent of whether the tower is trayed or
packed.
• Thus, the method for determining the minimum absorbent liquid or stripping
vapor flow rates in a packed column is identical to that for trayed towers, as
presented in section 6.3 and illustrated in Figure 6.10.
• The rate of mass transfer for absorption or stripping can be expressed in terms of mass-transfer coefficients for each phase. c
kc unit  kgmol/m2. s = kc kgmol/m3, kc is m/s

nA = kc a kc a  ka  unit  kgmol/s = ka kgmol/m3, ka is 1/s volumetric mass transfer coefficient

• Coefficients, k, based on a unit area for mass transfer could be used, but the area for interfacial mass transfer in a packed
bed is difficult to determine. Accordingly, it is common to use volumetric mass-transfer coefficients, ka (1/s), where the
variable a represents the interfacial area for mass transfer between phases per unit volume of packed bed. In this case nA is
kgmol/s.m3 of packed bed.
• At steady state, in the absence of chemical reactions, the rate of solute mass transfer
through the gas-phase film must equal the rate through the liquid film.

where kya is the volumetric mass-transfer coefficient for the gas phase and kxa is the
volumetric mass-transfer coefficient for the liquid phase.

• Thus, as shown in Figure 6.31, a straight

line of slope − kx a∕ky a, drawn from the
operating line at the bulk composition
point (y, x), intersects the equilibrium
curve at (yI , xI).
Henry’s law From Chapter 3

• At the interface, cAI and pAI are in equilibrium.

• Applying a version of Henry’s law,

• Different forms of Henry’s law are found in the literature. They include

• When a Henry’s law constant, HA, is given without citing the defining equation, the equation can be determined from the
units of the constant.

• For example, if the constant has the units of atm or atm/mole fraction, Henry’s law is

• If the units are mol∕L-mmHg, Henry’s law is

• Equilibrium at the interface can now be expressed in terms of a K-value for vapor–liquid equilibrium instead of as a Henry’s
law constant. Thus,

• Alternatively, fictitious concentrations and overall mass transfer coefficients can be used with mole-fraction driving forces.
Thus, and . If the two values of KA are equal,
• The slope −kxa∕kya determines the relative resistances of the two phases to mass transfer.
• In Figure 6.31, the distance AE is the gas-phase driving force (y − yI), while AF is the liquid-phase driving force (xI − x).
• If the resistance in the gas phase is very low, yI is approximately equal to y. Then, the resistance resides entirely in the
liquid phase. This occurs in the absorption of a slightly soluble solute in the liquid phase (a solute with a high K-value, e.g.,
CO2) and is referred to as a liquid-film controlling process.
• Alternatively, if the resistance in the liquid phase is very low, xI is nearly equal to x. This occurs in the absorption of a very
soluble solute in the liquid phase (a solute with a low K-value, e.g., NH3) and is referred to as a gas-film controlling process.
• It is important to know which of the two resistances is controlling, so that its rate of mass transfer can be increased by
promoting turbulence and/or increasing the dispersion of the controlling phase.
• The composition at the interface between two phases is difficult to measure, so overall
volumetric mass-transfer coefficients are defined in terms of overall driving forces
between the two phases. Using mole fractions,

where Kya and Kxa are the overall volumetric mass-transfer coefficients, and, as shown in
Figure 6.31 and previously discussed in §3.7, y∗ is the fictitious vapor mole fraction in
equilibrium with the bulk liquid mole fraction, x, and x∗ is the fictitious liquid mole
fraction in equilibrium with the bulk vapor mole fraction, y.
By combining (6-72) to (6-74), overall coefficients can be expressed in terms of separate
phase coefficients as follows:

• From Eq. (6-72) and (6-74), the following equation is obtained:

(6-74b)

• Solving Eq. (6-73) for y, , gives:

y = – (x – xI) (kxa / kya) + yI (6-74c)

• Substituting Eq. (6-74c) in Eq. (6-74b) to eliminate y, and solve for , gives
• Similarly for
• From Eq. (6-72) and (6-74), the following equation is obtained:
(6-74d)
• Solving Eq. (6-73) for x, , gives:

x = – (y – yI) (kya / kxa) + xI (6-74e)

• Substituting Eq. (6-74c) in Eq. (6-74b) to eliminate y, and solve for , gives

• Now, we have,

By Triangular Similarity
= yI / xI
In Figure 6.31, for dilute solutions, when the equilibrium curve is a nearly straight line
through the origin,
= xI / yI

where K is the K-value for the solute. Combining (6-75) with (6-77), and (6-76) with
(6-78), gives

Same results we will obtain if we and

follow up slides 29-34 of Chapter 3
• Column packed height is determined from the overall gas-phase coefficient, Kya, when the liquid has a strong affinity for the
solute, so that the resistance to mass transfer is mostly in the gas phase.
• This is analogous to a trayed tower, for which the tray-efficiency analysis is commonly based on . In the
countercurrent-flow absorption column in Figure 6.32 for a dilute system, a differential material balance for a solute being
absorbed in a differential height of packing dl gives

where AT is the inside cross-sectional area of the tower.

In integral form, with nearly constant factors placed outside the integral, (6-81) becomes

y - dy
• Solving for the packed height, - dy dx

or y
Chilton and Colburn [43] suggested that the RHS of (6-83) be written as
the product of two terms:

• Comparing (6-84) to (6-68) shows that HOG is analogous to HETP, just as NOG is analogous to Nt.
• HOG is the overall height of a (gas) transfer unit (HTU). Experimental data show that HTU varies less with V than does Kya.
The smaller the HTU, the more efficient the contacting. NOG is the overall number of (gas) transfer units (NTU). It
represents the overall change in solute mole fraction divided by the average mole-fraction driving force. The larger the NTU,
the greater the time or area of contact required. Equation (6-86) was integrated by Colburn [44], who used a linear
equilibrium, y∗= Kx, to eliminate y∗ and a linear, solute material-balance operating line, (6-70), to eliminate x, yielding

• Using (6-88) and (6-85), the packed height, lT , can be determined from (6-84). However, (6-88) is a very sensitive
calculation when A < 0.9.
• NTU (NOG) and HTU (HOG) are not equal to the number of equilibrium
stages, Nt, and HETP, respectively, unless the operating and
equilibrium lines are straight and parallel.
• Otherwise, NTU is greater than or
less than Nt, as shown in Figure
6.33, for the case of absorption.
When the operating and equilibrium
lines are straight but not parallel,
• Although most applications of HTU and NTU are based on (6-84) to (6-86) and (6-88), alternative groupings have been used, depending on
the driving force for mass transfer and whether the resistance to mass transfer is mostly in the gas or the liquid to which H OL and NOL apply.
• These groupings are summarized in Table 6.5. This table includes driving forces based on partial pressures, p; mole ratios, X, Y;
concentrations, c; and mole fractions, x, y.
• For later reference, Table 6.5 also provides groupings for UMD when solute concentration is not dilute.
• In the previous development, gas and liquid mass-transfer coefficients are based on a mole-fraction driving force.
• However, other driving forces are often used for packed columns, including partial pressure and mole ratio for a gas, and
concentration and mole ratio for a liquid. In those cases, (6-72) for the rate of component mass transfer, r, in moles per unit
time per volume of packed bed, is written in terms of these other driving forces as

for the gas phase

for the liquid phase

• Occasionally, a concentration driving force is also used for the gas phase, with the same kG symbol.
• When a concentration driving force is used for either the gas or the liquid, the units for the mass-transfer coefficient are
length per unit time (m/s).
• It is frequently necessary to convert a mass-transfer coefficient based on one driving force to another coefficient based on
a different driving force.
• Table 3.12 provides the relationships among the different coefficients.
• A partial-pressure driving force is common for the gas phase, with kG being the symbol for the mass-transfer coefficient.
• A molar concentration driving force is common for the liquid phase, with kL being the symbol for the mass-transfer
coefficient.

• In the three examples that follow, the first and second demonstrate processes for which the gas phase resistance is
dominant, and the third deals with a system for which both gas and liquid resistances exist
 Note From Chapter 3
• When using handbook or literature correlations to estimate mass-transfer coefficients, it is important to determine which
coefficient (kp, kc, ky, or kx) is correlated, because often it is not stated. This can be done by checking the units or the form
of the Sherwood or Stanton numbers.
• The various coefficients are related by the following expressions, which are summarized in Table 3.12.
6.8 PACKED-COLUMN LIQUID HOLDUP, DIAMETER, FLOODING, PRESSURE DROP, AND MASS-TRANSFER EFFICIENCY
• The values of volumetric mass-transfer coefficients and HTUs depend on gas and liquid velocities.
• These, in turn, depend on gas and liquid flow rates and column diameter.
• The estimation of column diameter for a given system, packing, and operating conditions requires the consideration of liquid
holdup, loading, flooding, and pressure drop
6.8.1 Liquid Holdup
6.8.2 Pressure Drop, Flooding, and Column Diameter
6.8.3 Mass-Transfer Efficiency

Note: For liquid holdup, pressure drops, flooding, column diameter and mass transfer efficiency, correlations and models have
been developed for packed bed absorbers and strippers. For these and design considerations and knowledge see the textbook
and open and patent literature.
6.9 REACTIVE (CHEMICAL) ABSORPTION
• In physical absorption, no significant chemical reactions occur between the solute being absorbed from the gas and the absorbent.
• Common absorbents in physical absorption, as listed in Table 6.1, are water and hydrocarbon oils.
• Often, the absorption step is followed by stripping or distillation to recover the solute and recycle the absorbent.
• Often, especially when the solute is an acid gas, it is more efficient to use chemical absorption.
• Acid gases include CO2, H2S, N2O, SO2, SO3, HCl, and HCN, for which a chemical reaction with the absorbent occurs in the liquid phase.
• If the absorbent is an aqueous solution of a strong base, such as NaOH or KOH, ionic reactions occur that are irreversible and either fast or
instantaneous.
• For example, if the solute is CO2 and the absorbent is aqueous sodium hydroxide, the overall reaction is

where the salt, Na2CO3, is soluble in water. The liquid leaving the absorber may require further treatment
• Acid gases, particularly CO2 and H2S, are also removed by aqueous amine solutions, including monoethanolamine (MEA), diethanolamine
(DEA), diethyleneglycol (DEG), and triethyleneglycol (TEG), with slower, but reversible, chemical reactions occurring in the liquid phase.
• In this case, the reactions are reversed downstream of the absorber in a subsequent operation to enable the recovery of the acid gases and
the recycling of the aqueous amine absorbent.
• For both types of chemical absorption, fast chemical reactions increase the rate of absorption by reducing the mass-transfer resistance of
the liquid phase, increasing the absorption capacity of the solvent used in the absorbent, and increasing the selectivity for certain
components of the gas.
• In some cases, a hazardous chemical is converted to a safe one.

• Gas–liquid reactions in absorbers are treated in a number of books, including Astarita [51], Danckwerts [52], Sherwood, Pigford, and Wilke
[53], Shah [54], and Doraiswamy and Sharma [55].
• Here, the subject of reactive absorption is introduced by considering an irreversible reaction occurring mainly in the liquid film adjacent to
the two-phase interface in a packed tower.
• This subject can be extended to more difficult cases involving reversible reactions by the methods in Chapter 12, using process simulators
• Assume the gas fed to an absorber contains component A, which is soluble in the
countercurrently flowing nonvolatile absorbent. Except for A, the feed gas is insoluble in the
absorbent. The absorbent includes component B, which reacts irreversibly with A to produce
nonvolatile products that are soluble in the absorbent by the reaction

• Composition profiles for A and B in the fluid films for three cases are shown in the vicinity of
the gas–liquid interface, I, in Figure 6.44, where the driving forces for mass transfer are taken as
partial pressure in the gas phase and molar concentrations in the liquid phase.
• In all three cases, the amount of B is in excess of that required for complete reaction of A.
Physical equilibrium is assumed at the gas–liquid interface I.
Case (a) – Infinitely Slow Reaction
• No reaction occurs anywhere in the absorber.
• The absorption is only physical, and the rate of absorption of A depends only on the mass-
transfer resistances in the gas and liquid phases.
• The rate of absorption, in terms of resistances, is obtained by combining (6-105), (6-108), and
(6-109) to give the following equation, where all terms without a component subscript apply
to component A:

• If A is very soluble in the liquid phase, the rate of absorption is controlled by the gas phase;
otherwise, the liquid phase controls.
Case (b) – Instantaneous reaction
• The chemical reaction between A and B is instantaneous at the phase interface.
• Accordingly, the partial pressure and concentration of A at the phase interface is zero.
• The only resistance to the rate of mass transfer of A to the liquid phase is in the gas phase, as given
by the following modification of (6-128):

Case (c) – Intermediate reaction rate

• The reaction occurs at an intermediate rate, both in the liquid film, adjacent to the phase interface,
and in the bulk liquid in the presence of excess B.
• The rate of mass transfer of A depends on the resistances in the two phases and the rate of the
chemical reaction.
• If it is assumed that the reaction is a pseudo first-order reaction in B, (6-128) is modified, according
to the treatment by Hatta, as described by Sherwood et al. [53], to give

This term accounts for the

where the mass-transfer resistance in the gas phase is pseudo first-order
unchanged, but the mass-transfer resistance in the liquid phase reaction in B that occurs in
is enhanced by the chemical reaction occurring in the liquid the bulk liquid, where hL
is the fractional
film according to an enhancement factor, E, which is > 1.
volumetric liquid holdup.
• If the chemical reaction is fast, and occurs only in the liquid film, and if the
concentration of B drops to zero at the phase interface, I, the third term in the
denominator of (6-130) is omitted
• As shown in Figure 6.45, the enhancement factor is a function of the Hatta number, , where is the ratio of the maximum
possible chemical conversion in the liquid film to the maximum mass-transfer rate through the film.
• For the pseudo first-order reaction, where k is the reaction rate constant,

• The enhancement factor is also a function of the parameter Ei, which is the enhancement
factor for an instantaneous chemical reaction, as given by

• If the reaction is instantaneous, the only resistance is that of the gas phase. For a case
of the absorption of CO2, Sherwood et al. [53] show that the overall mass-transfer
coefficient, KG, for absorption in 2-N KOH is 76 times greater than for absorption in
water
• When the chemical reaction is instantaneous at the phase interface, the
required number of equivalent stages or the number of transfer units decreases.
Because the concentration and partial pressure of the solute become zero, the
equilibrium curve is a horizontal line at an ordinate value of zero:

• For a 10-fold decrease, the partial pressure of the solute NOG = 2.3. For a 100-fold decrease, NOG = 4.6