United States Air Pollution Training Institute EPA 450/2-82-020
Environmental Protection MD 20 March 1984
Agency Environmental Research Center
Research Triangle Park, NC 27711
vvEPA APTI
Course SI:412C
Wet Scrubber Plan Review
Self-instructional
Guidebook
CHEEPS. INC.
EPA LIBRARY SERVICES RTP NC
TECHNICAL DOCUMENT COLLECTION
-------
United States
Environmental Protection
Agency
Air Pollution Training Institute
MD 20
Environmental Research Center
Research Triangle Park, NIC 27711
ePA 450^82-020
|Vl9rch
Air
APTI
Course SI:412C
Wet Scrubber Plan Review
Self-instructional
Guidebook
Written by:
Gerald T. Joseph. P.E.
Oavid S. Beachler
Instructional Design:
Marilyn M. Peterson
Edited by:
Cathy O. Warren
Northrop Services, Inc.
P.O. Box 12313
Research Triangle Park. NC 27709
Under Contract No.
68-02-3573
EPA Project Officer
R. E. Town send
United States Environmental Protection Agency
Office of Air, Noise, and Radiation
Office of Air Quality Planning and Standards
Research Triangle Park, NC 27711
-------
Notice
This is not an official policy and standards document. The opinions and selections are
those of the authors and not necessarily chose of the Environmental Protection Agency.
Every attempt has been made to represent the present state of the art as well as subject
areas still under evaluation, Any mention of products or organizations does not. con-
stitute endorsement by the United States Environmental Protection Agency.
The authors request that any material abstracted from this manual be appropriately
referenced as a matter of professional courtesy in the following manner:
Joseph, G, T. and Beachler, D. S. 1984. Wet Scrubber Plan Review—Self Instruc-
tional Guidebook. APTI Course SI:412C, EPA 450/2-82-020.
ii
-------
Table of Contents
Page
Course Introduction v
Lesson 1 Introduction to Wet Scrubbers 1-1
Lesson Goal and Objectives 1 -1
Introduction 1-1
Particle Collection 1-3
Gas Collection 1-7
Categorizing Wet Scrubbers 1-9
Lesson 2 Design Features of Wet Scrubbers 2-1
Lesson Goal and Objectives 2 1
Designing Wet Scrubbers 2 1
Scrubber Components 2-1
Lesson 3 Gas-Phase Contacting Scrubbers 3-1
Lesson Goal, and Objectives 3-1
Introduction 3-1
Venturi Scrubbers 3-2
Plate Towers 3-15
Orifice Scrubbers 3-22
Lesson 4 Liquid-Phase Contacting Scrubbers 4-1
Lesson Goal and Objectives 4-1
Introduction 4-1
Spray Towers 4 1
Ejector Venturis 4 6
Lesson 5 Wet-Film Scrubbers . 5-1
Lesson Goal and Objectives 5-1
Introduction 5-1
Gas Collection 5-2
Tower Designs 5 2
Packing Material j6
Exhaust Gas Distribution 5-8
Liquid Distribution. 5-8
Maintenance Problems 5-10
Summary ,,., 5-14
Lesson 6 Combination Devices—Liquid-Phase and Gas-Phase'
Contacting Scrubbers 6-1
Lesson Goal and Objectives 6-1
Introduction 6-1
Cyclonic Spray Scrubbers 6-1
Mobile-Bed Scrubbers 6-5
Baffle Spray Scrubbers o 11
Mechanically Aided Scrubbers 6 12
iii
-------
Page
Lesson 7 Equipment Associated with Scrubbing Systems 7-1
Lesson Goal and Objectives 7-1
Introduction 7-1
Transport Equipment for Exhaust Gases and Scrubbing Liquids 7-1
Conditioning Equipment for Exhaust Gases 7-5
Construction Materials 7-6
Monitoring Equipment 7-8
Lesson 8 Wet Flue Gas Desulfurization Systems 8-1
Lesson Goal and Objectives 8-1
Introduction 8-1
Nonregenerable FGD Processes 8-6
Regenerable FGD Processes 8-26
Summary 8-31
Lesson 9 Design Review of Scrubbers Used for Particulate Pollutants 9-1
Lesson Goal and Objectives 9-1
Introduction 9-1
Wet Scrubbers Used to Remove Particles 9-2
Review of Design Criteria for Permits 9-19
Summary 9-19
Lesson 10 Design Review of Absorbers Used for Gaseous Pollutants 10-1
Lesson Goal and Objectives 10-1
Introduction 10-1
Absorption 10-2
Solubility 10-2
Absorber Design 10-9
Review of Design Criteria for Permits 10-42
Summary 10-43
IV
-------
Course Introduction
Description
This course is designed for engineers and other technical persons responsible for reviewing
plans for the installation of wet scrubbers. This course focuses on review procedures for wet
scrubbers used to reduce particulate and gaseous emissions from industrial sources. Major
topics related to wet scrubbers include the following:
• General description
• Particle collection and absorption theory
• Estimating collection efficiency
• Components
• Use in flue gas desulfurization (FGD)
• Operation and maintenance problems.
Course Goal and Objectives
Course Goal
To familiarize you with wet scrubbers — how they operate, their common operation and
maintenance problems, and the review steps for evaluating their installation plans.
Course Objectives
At the end of the course, you should be able to—
1. recognize various scrubbers and briefly describe their operation,
2. briefly describe the mechanisms for particle collection and gas absorption in a scrubber,
3. recognize which scrubbers are used mainly for particle collection and which are used
mainly for gaseous pollutant removal,
4. briefly describe four FGD systems used for removing sulfur dioxide emissions from
boilers,
5. list three key design parameters affecting particle and gaseous pollutant removal, and
6. recognize typical operation and maintenance problems associated with each wet
scrubber.
Requirements for Successful Completion
In order to receive 4.0 Continuing Education Units (CEUs) and a certificate of course comple-
tion, you must achieve a final examination grade of at least 70%.
-------
Materials
Reading
This text — supplementary reading materials are not required.
Using the Guidebook
This book directs your progress through the course. Ten lessons describe wet scrubbers and
how they are used to control particulate and gaseous emissions.
There is a review exercise at the end of each lesson. To complete an exercise, place a piece
of paper across the page, covering the questions below the one you are answering. After
answering the question, slide the paper down to uncover the next question. The answer for
the first question will be given on the right side of the page, separated by a line from the
second question, as shown here. All answers to review questions will appear below and to the
right of their respective questions. The answer will be numbered to match the question.
Please do not write in this book. Complete each review exercise in the lessons. If you are
unsure about a question or answer, review the material preceding the question. Then proceed
to the next section.
Review Exercise
1. Question I on l<>
nil i cllo ylloniilic
2. Questionoli oul
li iilnoiiyic o
1. Answer
llllO
3. Question > hi lot
nil i cllo ylloii
2. Answer
Lesson Content
• Lesson goal and objectives
• Text of lesson
• Review exercise and review exercise answers
The material contained in Lessons 9 and 10 is a review of the design theory for particle
scrubbers and gaseous pollutant absorbers. Much of this material was covered in APTI
Courses 413, Control of Particulate Emissions, and 415, Control of Gaseous Emissions.
vi
-------
However, these lessons provide a good review of the design theory and equations used to
estimate collection efficiency, liquid-injection rates, absorber diameter, and the number of
plates used in a plate tower. Material in all 10 lessons is covered in the final examination.
Instructions for Completing
the Final Examination
Contact the Air Pollution Training Institute if you have any questions about the course or
when you are ready to receive a copy of the final examination.
After completing the final exam, return it and the answer sheet to the Air Pollution Train-
ing Institute. The final exam grade and course grade will be mailed to you.
Air Pollution Training Institute
Environmental Research Center
MD 20
Research Triangle Park, NC 27711
vn
-------
Lesson 1
Introduction to Wet Scrubbers
Lesson Goal and Objectives
Goal
To familiarize you with the variables that affect particle and gas collection in wet scrubbers.
Objectives
Upon completing this lesson, you should be able to —
1. list four advantages and disadvantages of using wet scrubbers to collect panicles and
gases rather than using other air pollution control devices,
2. describe the two most important mechanisms for collecting particles in wet scrubbers,
3. name three process variables that affect particle collection in a wet scrubber,
4. describe the process of absorption, and
5. list three conditions that will enhance the absorption process.
Introduction
Wet scrubbers are air pollution control devices that use liquid to remove particles or gases
from industrial exhaust streams. The dirty exhaust stream is brought into contact with the
liquid by spraying it with the liquid, by forcing it through a pool of liquid, or by some other
contact method. When wet scrubbers are used for removing particles, the particles are cap-
tured by and incorporated into liquid droplets. These droplets must then be separated from
the clean exhaust stream. When wet scrubbers are used for removing gases, the gases are
dissolved or absorbed by the liquid.
The advantages or disadvantages of using a wet scrubber instead of some other control
device depend on the pollutant (gas or particle) to be controlled. Wet collectors, baghouses,
or electrostatic precipitators can be used when collecting small particles at a high efficiency
(>95%) is necessary. When only large particles are to be removed, either a low-energy scrub-
ber or a cyclone can be used. Choosing the "best" collection system depends on many factors.
Often no obvious choice is best. Table 1-1 contains some of the advantages and disadvantages
of using a wet collector to remove particulate and gaseous emissions.
1-1
-------
Table 1-1. Relative advantages and disadvantages of wet scrubbers.
Advantages
Disadvantages
Small space requirements
Scrubbers reduce the temperature and
volume of the unsaturated exhaust stream.
Therefore, vessel sizes, including fans and
ducts downstream, are smaller than those
of other control devices. Smaller sizes result
in lower capital costs and more flexibility
in site location of the scrubber.
No secondary dust sources
Once particles are collected, they cannot
escape from hoppers or during transport.
Handles high-temperature, high-humidity
gas streams
No temperature limits or condensation
problems can occur as in baghouses or
ESPs.
Minimal fire and explosion hazards
Various dry dusts are flammable. Using
water eliminates the possibility of
explosions.
Ability to collect both gases and particles
Corrosion problems
Water and dissolved pollutants can form
highly corrosive acid solutions. Proper con-
struction materials are very important.
Also, wet-dry interface areas can result in
corrosion.
High power requirements
High collection efficiencies for particles are
attainable only at high pressure drops,
resulting in high operating costs.
Water-disposal problems
Settling ponds or sludge clarifiers may be
needed to meet waste-water regulations.
Difficult product recovery
Dewatering and drying of scrubber sludge
make recovery of any dust for reuse very
expensive and difficult.
Meteorological problems
The saturated exhaust gases can produce a
wet, visible steam plume. Fog and precipi-
tation from the plume may cause local
meteorological problems.
For gaseous pollutant removal, the choice of the control device depends mainly on the type
of gaseous pollutant to be controlled. In choosing a system to control organic vapors, the
choice of control is among wet scrubbers, adsorbers, thermal oxidizers (incinerators), or con-
densers; to control most inorganic gases (HC1, H2S, HF, and S02), a wet scrubber is usually
the primary control device. If the exhaust stream contains both particles and gases, wet scrub-
bers are generally the only air pollution control device used to remove both pollutants. One
exception is using a baghouse or an electrostatic precipitator (ESP) with a spray dryer in a dry
S02 scrubbing system.
Wet scrubbers can achieve high removal efficiencies for either particles or gases and, in
some instances, can achieve a high removal efficiency for both pollutants in the same system.
However, in many cases, the best operating conditions for particle collection are the poorest
for gas removal. In general, obtaining high simultaneous gas and particle removal efficiencies
requires that one of them be easily collected (i.e., that the gases are very soluble in the liquid
or that the particles are large and readily captured). Wet scrubbers have been used in a vari-
ety of industries such as acid plants, fertilizer plants, steel mills, asphalt plants, and large
power plants.
This lesson will examine operating variables that influence scrubber performance for both
particle and gas collection.
1-2
-------
Review Exercise
1. True or False? Particles and gases are absorbed in the
scrubbing liquid.
2. In choosing a control device for high collection efficiency
of small particles, wet collectors are compared to
/' '' or V. "v ^
1. False
Only gases are absorbed.
3. When choosing a device for organic-vapor collection, wet
collection is compared ro or Oci<-o<~c
2. baghouses (or)
electrostatic precipitators
4. In general, high removal rates for both particles and gases
in the same scrubber are obtained by
a. the use of large amounts of water.
b. having gases that are highly soluble and/or particles
that are relatively large. U .
c. the use of extremely high pressure drops. br
d. a reagent added to the water.
3. incineration (or)
adsorption
4. b. having gases that are
highly soluble and/or
particles that are
relatively large.
Particle Collection
Wet scrubbers capture relatively small dust particles with large liquid droplets. Droplets are
produced by injecting liquid at high pressure through specially designed nozzles, by aspirating
the panicle-laden gas stream through a liquid pool, or by submerging a whirling rotor in a
liquid pool. These droplets collect particles by using one or more of several collection
mechanisms. These mechanisms —impaction, direct interception, diffusion, electrostatic
attraction, condensation, centrifugal force, and gravity —are listed in Table 1-2. However,
impaction and diffusion are the two primary ones.
1-3
-------
Table 1-2. Particle collection mechanisms for wet scrubbing systems.
Mechanism
Explanation
Impaction
Particles too large to follow gas streamlines around a droplet collide
with it.
Diffusion
Very tiny particles move randomly, colliding with droplets because they
are confined in a limited space.
Direct interception
An extension of the impaction mechanism. The center of a particle
follows the streamlines around the droplet, but a collision occurs if the
distance between the panicle and droplet is less than the radius of the
particle.
Electrostatic attraction
Particles and droplets become oppositely charged and attract each
other.
Condensation
When hot gas cools rapidly, particles in the gas stream can act as con-
densation nuclei and, as a result, become larger.
Centrifugal force
The shape or curvature of a collector causes the gas stream to rotate in
a spiral motion, throwing larger particles toward the wall.
Gravity
Large particles moving slowly enough will fall from the gas stream and
be collected.
Impaction
In a wet scrubbing system, dust particles will tend to follow the streamlines of the exhaust
stream. However, when liquid droplets are introduced into the exhaust stream, particles can-
not always follow these streamlines as they diverge around the droplet (Figure 1-1). The par-
ticle's mass causes it to break away from the streamlines and impact on the droplet. Impac-
tion is the predominant collection mechanism for scrubbers having gas stream velocities
greater than 0.3 m/s (1 ft/sec) (Perry 1973). Most scrubbers do operate with gas stream
velocities well above 0.3 m/s. Therefore, at these velocities, particles having diameters greater
than 1.0 /xm are collected by this mechanism.
Gas streamlines
Droplet
Particle
Figure 1-1. Impaction.
1-4
-------
As the velocity of the particles in the exhaust stream increases relative to the liquid droplets'
velocity, impaction increases. Impaction also increases as the size of the liquid droplet
decreases. This is because there will be more droplets (for the same amount of liquid) within
the vessel, consequently increasing the likelihood that the particles will impact on the droplets.
Diffusion
Very small panicles (less than 0.1 /»m in diameter) experience random movement in an
exhaust stream. These particles are so tiny that they are bumped by gas molecules as they
move in the exhaust stream. This bumping, or bombardment, causes them to first move one
way and then another in a random manner, or diffuse, through the gas. This irregular
motion can cause the particles to collide with a droplet and be collected (Figure 1-2). Because
of this, in certain scrubbers, the removal efficiency of particles smaller than 0.1 jxxn can
actually increase.
Gas streamlines
Droplet
Particle
Figure 1-2. Diffusion.
The rate of diffusion depends on relative velocity, particle diameter, and liquid-droplet
diameter. As for impaction, collection due to diffusion increases with an increase in relative
velocity (liquid- or gas-pressure input) and a decrease in liquid-droplet size. However, collec-
tion by diffusion increases as particle size decreases. This mechanism enables certain scrubbers
to effectively remove the very tiny particles. In the particle size range of approximately 0.1 to
1.0 nm, neither of these two dominates. Particles in this size range are not collected as effi-
ciently as are either larger particles collected by impaction or smaller particles collected by
diffusion.
Other Collection Mechanisms
In recent years, some scrubber manufacturers have designed scrubbers to use other collection
mechanisms such as electrostatic attraction and condensation to enhance particle collection
without increasing power consumption. Other mechanisms such as gravity, centrifugal force,
and direct interception slightly affect particle collection.
1-5
-------
Review Exercise
1. What are two primary mechanisms used for collecting par-
ticles in a scrubber?
a. impaction and diffusion
b. direct interception and diffusion •/-
c. impaction and condensation
d. direct interception and gravity
2. is/are the predominant collection mecha-
nism^) for panicles larger than 1.0 nm traveling faster
than 0.3 m/s (1 ft/sec).
a. Impaction . ,
b. Diffusion
c. Direct interception
d. all of the above
1. a. impaction and
diffusion
3. For very small particles, below 0.1 ^m in diameter,
is/are the predominant collection
mechanism(s) in wet collection.
a. impaction j
b. diffusion O
c. gravity
d. all of the above
2. a. Impaction
4. Collection efficiency for particles captured by the
impaction mechanism increases as the
a. particles' velocity in the exhaust stream increases
relative to the liquid droplets' velocity.
b. particle size decreases below 0.1 jim in diameter.
c. liquid-droplet size increases.
3. b. diffusion
5. Collection efficiency for particles captured by diffusion
decreases/increases as the size of the particle decreases.
4. a. particles' velocity in
the exhaust stream
increases relative to
the liquid droplets'
velocity.
5. increases
1-6
-------
Gas Collection
The process of dissolving gaseous pollutants in a liquid is referred to as absorption. Absorption
is a mass-transfer operation. Mass transfer can be compared to heat transfer in that both
occur because a system is trying to reach equilibrium conditions. For example, in heat
transfer, if a hot slab of metal is placed on top of a cold slab, heat energy will be transferred
from the hot slab to the cold slab until both are at the same temperature (equilibrium). In
absorption, mass instead of heat is transferred as a result of a concentration difference, rather
than a heat-energy difference. Absorption continues as long as a concentration differential
exists between the liquid and the gas from which the contaminant is being removed. In
absorption, equilibrium depends on the solubility of the pollutant in the liquid.
To remove a gaseous pollutant by absorption, the exhaust stream must be passed through
(brought in contact with) a liquid. Figure 1-3 illustrates the three steps involved in absorption.
In the first step, the gaseous pollutant diffuses from the bulk area of the gas phase to the gas-
liquid interface. In the second step, the gas moves (transfers) across the interface to the liquid
phase. This step occurs extremely rapidly once the gas molecules (pollutant) arrive at the
interface area. In the third step, the gas diffuses into the bulk area of the liquid, thus making
room for additional gas molecules to be absorbed. The rate of absorption (mass transfer of the
pollutant from the gas phase to the liquid phase) depends on the diffusion rates of the pollu-
tant in the gas phase (first step) and in the liquid phase (third step).
Gaseous
pollutant
Figure 1-3. Absorption.
To enhance gas diffusion and, therefore, absorption:
1. provide a large interfacial contact area between the gas and liquid phases,
2. provide good mixing of the gas and liquid phases (turbulence), and
3. allow sufficient residence, or contact, time between the phases for absorption to occur.
Two of these three gas-collection mechanisms, large contact area and good mixing, are also
important for particle collection. The third factor, sufficient residence time, works in direct
opposition to efficient particle collection. To increase residence time, the relative velocity of
1-7
-------
the gas and liquid streams must be reduced. Therefore, achieving a high removal efficiency
for both gaseous and particulate pollutants is extremely difficult unless the gaseous pollutant is
very soluble in the liquid.
As previously mentioned, a very important factor affecting the amount of a pollutant that
can be absorbed is its solubility. Solubility governs the amount of liquid (liquid-to-gas ratio)
required and the necessary contact time. More soluble gases require less liquid. Also, more
soluble gases will be absorbed faster. Solubility is a function of both the temperature and, to a
lesser extent, the pressure of the system. As temperature increases, the amount of gas that can
be absorbed by a liquid decreases. From the ideal gas law: as temperature increases, the
volume of a gas also increases; therefore, at a higher temperature, gas volume increases and
less gas is absorbed. For this reason, some absorption systems use inlet quench sprays to cool
the incoming exhaust stream, thereby increasing absorption efficiency. Pressure affects the
solubility of a gas in the opposite manner. When the pressure of a system is increased, the
amount of gas absorbed generally increases.
Review Exercise
1. In absorption, gaseous pollutants are in a
liquid.
2. Absorption occurs herause of a difference
between the gas phase and liquid phase.
a. heat
b. mass ,/^,'
c. concentration
d. weight
1. dissolved
3. Which of the following would not enhance the absorption
process?
a. providing a large contact area between the gas and
liquid phases /^-
b. providing a turbulent mixing of the phases
c. increasing the gas velocity relative to the liquid velocity
d. allowing long contact, or residence, time
2. c. concentration
4. True or False? The solubility of the gaseous pollutant in
the liquid will affect the required liquid-to-gas ratio ,
of the system.
3. c. increasing the gas
velocity relative to the
liquid velocity
5. Which of the following reduces the solubility of gas in
a liquid?
a. increased temperature
b. increased pressure W
4. True
1-8
-------
5. a. increased temperature
As the temperature
increases, the amount
of gas that can be
absorbed decreases
because the gas
expands.
Categorizing Wet Scrubbers
Since wet scrubbers vary greatly in complexity and method of operation, devising categories
into which all of them would neatly fit is extremely difficult. Scrubbers for particle collection
are usually categorized by the gas-side* pressure drop of the system. They are:
• low-energy scrubbers having pressure drops of less than 12.7 cm (5 in.) of water,
• medium-energy scrubbers having pressure drops between 12.7 and 38.1 cm (5 and 15 in.)
of water, and
• high-energy scrubbers having pressure drops greater than 38.1 cm (15 in.) of water.
However, most scrubbers operate over a wide range of pressure drops, depending on their
specific application, thereby making this type of categorization difficult.
Another way to categorize scrubbers is by the manner in which the gas and liquid phases
are brought into contact. In this category, the scrubbers use power, or energy, from the gas
stream or the liquid stream, or they use some other method to bring the pollutant gas stream
into contact with the liquid. These categories are given in Table 1-4.
Table 1-4. Categories of wet collectors by energy source used for contact.
Wet collector
Energy source
used for contact
Gas-phase contacting
Liquid-phase contacting
Wet film
Combination
• Liquid phase and gas phase
• Mechanically aided
Gas stream
Liquid stream
Liquid and gas streams
Liquid and gas streams
Mechanically driven rotor
Another way to classify wet scrubbers is by their use —to primarily collect either particles or
gaseous pollutants, or both. Each of the wet collectors listed in Table 1-4 will be discussed in
this course, including various designs within each category, their operation, collection effi-
ciency, industrial applications, prominant maintenance problems, if any, and their primary
use.
* Gas-side pressure drop refers to the pressure difference, or pressure drop, that occurs as the exhaust gas is
pushed or pulled through the scrubber, disregarding the pressure that would be used for pumping or spraying
the liquid into the scrubber. In this manual, the terms pressure drop and gas-side pressure drop will be used
interchangeably.
1-9
-------
References
Beachler, D. S., andjahnke, J. A. October 1981. Control of particulate emissions. APTI
Course 413, EPA 450/2-80-066. U.S. Environmental Protection Agency.
Bethea, R. M. 1978. Air pollution control technology. New York: Van Nostrand Reinhold Co.
Joseph, G. T., and Beachler, D. S. December 1981. Control of gaseous emissions. APTI
Course 415, EPA 450/2-81-005. U.S. Environmental Protection Agency.
Perry, J. H., ed. 1973. Chemical engineers' handbook, 5th ed. New York: McGraw-Hill
Book Co.
Semrau, K. T. 1977. Practical process design of particulate scrubbers. Chem. Eng. 84:87-91.
1-10
-------
Lesson 2
Design Features of Wet Scrubbers
Lesson Goal and Objectives
Goal
To introduce you to the design features unique to wet scrubbers that enhance the collection of
air pollutants.
Objectives
Upon completing this lesson, you should be able to —
1. list at least six major components of a wet scrubber,
2. recognize three spray nozzle designs,
3. list at least four characteristics of spray nozzles,
4. list five remedies for plugged nozzles, and
5. describe the operation of three entrainment separators.
Designing Wet Scrubbers
Wet scrubbers are uniquely designed to enhance the collection of air pollutants. As discussed
in the last lesson, several process design variables affect particulate pollutant collection—most
importantly, particle size, particle velocity, and liquid-droplet size. For gaseous pollutant col-
lection, the pollutant must be soluble in the chosen scrubbing liquid. In addition, the system
must be designed to provide good mixing between the gas and liquid phases, and enough time
for the gaseous pollutants to dissolve. Another consideration for both particulate and gaseous
pollutant collection is the liquid-to-gas (L/G) ratio —the amount of liquid injected into the
scrubber per given volume of exhaust flow. Lastly, the system must be designed to remove
entrained mists, or droplets, from the cleaned exhaust gas stream before it leaves the stack.
Scrubber Components
Several components are used when designing scrubbers to provide gas-liquid contact and
separation. Spray nozzles are used to form droplets that, in turn, are used to capture
pollutants. Other components are used to enhance gas-liquid contact. These include venturi
throats, plates, baffles, packing, orifices, tangential openings, and mechanically driven rotors.
2-1
-------
These basic components are used by themselves or in combination in commercially available
scrubbers. These will be discussed in detail in the following lessons as they apply to a specific
scrubber design. Plastic pads, wire-mesh pads, blades, and cyclones are used to separate any
entrained droplets from the cleaned exhaust gas stream. Spray nozzles and entrainment
separators are found in some form in many scrubber systems. They will be discussed here and
mentioned later as they apply to each specific scrubber.
Spray Nozzles
Three different nozzle designs are used to produce a fine, cone-pattemed spray. In the
impingement nozzle (Figure 2-1), highly pressurized liquid passes through a hollow tube in the
nozzle and strikes a pin or plate at the nozzle tip. A very fine fog of tiny, uniform-sized
droplets approximately 25 to 400 fim in diameter is produced. Because there are no internal
parts in the nozzle, it will not plug as long as particles larger than the opening are filtered out
by a strainer. These nozzles are usually made of stainless steel or brass. In the solid cone
nozzle (Figure 2-2), liquid is forced over an insert to break it up into a cone of fine droplets.
Cones can be full, hollow, or square with spray angles from 15° to 140°. These nozzles can be
made of stainless steel, brass, alloys, Teflon®, and other plastic materials. The helical spray
nozzle (Figure 2-3), has a descending spiral impingement surface that breaks up the sprayed
liquid into a cone of tiny droplets. The cones can be full or hollow with spray angles from 50°
to 180°. There are no internal parts, which helps reduce nozzle plugging. These nozzles can
be made of stainless steel, brass, alloys, Teflon®, and other plastic materials.
Figure 2-1. Impingement nozzle. Figure 2-2. Solid cone nozzle. Figure 2-3. Helical spray nozzle.
2-2
-------
Different spray nozzles are appropriate for different scrubbing systems. Characteristics of the
nozzles and sprays include the following:
1. Droplet size —In general, scrubbers using sprays to provide gas-liquid contact (such as in
spray towers) require tiny, uniform-sized droplets to operate effectively. If the sprays are
used merely as a method of introducing liquid into the vessel (such as in packed towers),
then droplet size is not as critical.
2. Opening size —The actual opening in the nozzle will vary depending on the applications
and the amount of liquid required. Openings range from 0.32 to 6.4 cm (0.125 to
2.5 in.).
3. Spray pattern — Nozzles are available that produce sprays in a number of geometric
shapes such as square, fan, hollow cone, and full cone. Full-cone sprays are used to pro-
vide complete coverage of the areas sprayed.
4. Operating mechanism — Droplets can be produced by a number of methods such as
impinging the liquid on a solid surface or atomizing the liquid using air.
5. Power consumption — In general, the finer the liquid droplet, the higher the power
consumption.
Nozzle plugging is one of the most common malfunctions in wet scrubbers. Plugged nozzles
reduce the gas-liquid contact and can also result in scale buildup on, or heat damage to, the
scrubber parts formerly sprayed by the nozzle. Nozzle plugging can be most readily detected
by observing the liquid spray pattern; however, if the nozzles are not easily accessible, a
decrease in liquid flow is also a telltale sign (EPA 1982). Remedies include replacing the
nozzle with one that is more open, cleaning the nozzle frequently, filtering the scrubbing
liquid, or increasing the bleed rate and makeup water rates.
Another problem that can arise is reduced pressure in the spray header. This can cause a
reduction in the spray angle (area covered) and an increase in the size of droplets produced.
Entrainment Separators
As mentioned in Lesson 1, the pollutant must first be contacted with the liquid, then the
liquid droplets must be removed from the exhaust gas stream before it is exhausted to the
atmosphere. Entrainment separators, also called mist eliminators, are used to remove the
liquid droplets. Although the major function of an entrainment separator is to prevent liquid
carryover, it also performs additional scrubbing and recovers the scrubbing liquor, thus saving
on operating costs. Therefore, entrainment separators are usually an integral part of any wet
scrubbing system.
Entrained liquid droplets vary in size depending on how the droplets were formed. Droplets
that are torn from the body of a liquid are large (10 to 100 /im in diameter), whereas droplets
that are formed by a chemical reaction or by condensation are on the order of 5 jim or less in
diameter. Numerous types of entrainment separators are capable of removing these droplets.
Those most commonly used for air pollution control purposes are cyclonic, mesh-pad, and
blade separators.
The cyclonic (centrifugal) separator is a cylindrical tank with a tangential inlet or turning
vanes. The tangential inlet or turning vanes impart a swirling motion to the droplet-laden gas
stream. The droplets are thrown outward by centrifugal force to the walls of the cylinder.
Here they coalesce and drop down the walls to a central location and are recycled to the
absorber (Figure 2-4). These units are simple in construction, having no moving parts.
2-3
-------
Clean exhaust gas
Clean
exhaust gas
containing droplets
Figure 2-4. Cyclonic separator.
Therefore, they have few plugging problems as long as continuous flow is maintained. Good
separation of droplets 10 to 25 p.m in diameter can be expected. The pressure drop across the
separator is 10 to 15 cm (4 to 6 in.) of water for a 98% removal efficiency of droplets in the
size range of 20 to 25 /an. Cyclonic separators are commonly used with venturi scrubbers (see
Lesson 3).
In another design, wire or plastic is used to form mesh pads (Figure 2-5). These mesh-pad
separators are approximately 10 to 15 cm (4 to 6 in.) thick and fit across the entire diameter
of the scrubber. The mesh allows droplets to impact on the material surface, agglomerate
with other droplets, and drain off by gravity. The pad is usually slanted (no more than a few
degrees) to permit the liquid to drain off. Better than 95% collection of droplets larger than
3 jim is obtained with pressure drops of approximately 1.0 to 15 cm (0.5 to 6 in.) of water
(the pressure drop depends on depth and compaction of fibers). The disadvantage with mesh
pads is that their small passages are subject to plugging. Periodically spraying pads from both
below and above can remove some trapped material. However, spraying only from beneath
will drive entrapped material further into the mesh, necessitating removal of the pads for
cleaning or replacement (Schifftner 1979).
2-4
-------
Figure 2-5. Mesh-pad separator.
Blade separators can be of two types: chevron or impingement. In the chevron separator
(Figure 2-6), gas passes between the blades and is forced to travel is a zigzag pattern. The
liquid droplets cannot follow the gas streamlines, so they impinge on the blade surfaces,
coalesce, and fall back into the scrubber chamber or drain. Special features such as hooks and
pockets can be added to the sides of these blades to help improve droplet capture. Chevron
grids can be stacked or angled on top of one another to provide a series of separation stages.
Pressure drop is approximately 6.4 cm (2.5 in.) of water for capture of droplets as small as
5 ixm in diameter. Impingement separators (Figure 2-7) create a cyclonic motion because they
are similar in shape to the common house fan. As the gas passes over the curved blades, they
impart a spinning motion that causes the mist droplets to be directed to the vessel walls,
where they are collected. Pressure drop ranges from 5 to 15 cm (2 to 6 in.) of water.
Figure 2-7. Impingement blade separator
Figure 2-6. Chevron blade separator
-------
The most important diagnostic aid in monitoring separator performance is the pressure
drop. By measuring the pressure drop across the separator, the following problems can be
identified (Wechselblatt 1975):
• A sudden decrease in pressure drop at constant load indicates that the separators have
shifted out of place or are broken.
• An increase in pressure drop, even as little as 0.5 cm (0.2 in.) of water, is an indication
of material buildup in the separator.
Another diagnostic measurement is gas velocity. Gas velocity through the separator must be
kept below the maximum rate to avoid liquid reentrainment. Maximum velocities depend on
operating conditions and the physical properties of the exhaust gas and liquid streams. The
gas velocity should be kept below 3 m/s (10 ft/sec) for chevron separators, below 5 m/s
(15 ft/sec) for mesh pads, and below 8 m/s (27 ft/sec) for impingement blades to reduce
liquid carryover (Schifftner 1979). Table 2-1 summarizes some operating characteristics of
entrainment separators.
Table 2-1. Typical operational characteristics of entrainment separators.*
Type
Droplet size
collected at 99%
(/im)
Maximum gas velocity
Pressure drop
m/s
ft/sec
cm H.O
in. H.O
Mesh pads
S.O
5
15
1.0-15
0.5-6
Cyclone
10-25
20
65
10-15
4-6
Blades
Chevron
Impingement vane
35
20
3
8
10
27
6.4
5-15
2.5
2-6
•Note: Values in this table are given as a general guide only. The collection efficiency for various
droplet sizes depends on the gas velocity through the entrainment separators.
2-6
-------
Review Exercise
1. Identify the following spray nozzle used in a wet scrubbing
system.
W' v v'
•1 1
2. List five important characteristics of spray nozzles used
in wet scrubbing systems. ¦
1. impingement
3. True or False? Nozzle plugging is one of the most
common malfunctions in wet scrubbers.
' V
2. • opening size
• droplet size
• spray pattern
• operating mechanism
• power consumption
4. List five remedies for plugged nozzles.
3. True
5. Entrainment separators are used to
a. prevent liquid carryover.
b. recover scrubbing liquor.
c. perform additional scrubbing.
d. all of the above
4. • Replace nozzle with
one having a more
open design.
• Clean nozzles
frequently.
• Filter the scrubbing
liquor.
• Increase bleed rate.
• Increase makeup
water rate.
6. Cyclonic separators can remove liquid droplets as small
as in diamerer.
a. 0.01 fxm
b. 0.1 fxm \
c. 1.0 /tm U
d. 10.0 /im
5. d. all of the above
6. d. 10.0 fj.m
-------
7. True or False? Wire- or plastic-mesh pads are capable of
removing smaller droplets than are either cyclonic or blade
separators; however, they are also more susceptible to
plugging.
8 In creneral, wire-mesh pads should he ro
prevent plugging.
a. installed at a slant
b. sprayed from the bottom \
c. sprayed from the top ^
d. sprayed from the top and bottom
e. all of the above
7. True
8. d. sprayed from the top
and bottom
References
Calvert, S.; Jadmani, I. L.; Young, S.; and Stahlberg, S. October 1974. Entrainment
separators for scrubbers—initial report.
Environmental Protection Agency. September 1982. Control techniques for particulate
emissions from stationary sources—volume 1. EPA 450/3-81-005a.
National Asphalt Pavement Association. 1978. The maintenance and operation of exhaust
systems in the hot mix batch plant. Information Series 52, 2nd ed.
Schifftner, K. C. April 1979. Venturi Scrubber Operation and Maintenance. Presented for the
U.S. EPA Environmental Research Information Center, at Atlanta, GA.
Semrau, K. T. 1977. Practical process design of paniculate scrubbers. Chem. Eng. 84:87-91.
Wechselblatt, P. M. 1975. Wet scrubbers (particulates). Handbook for the operation and
maintenance of air pollution control equipment. F. L. Cross and H. E. Hesketh, eds.
Westport: Technomic Publishing Co., Inc.
2-8
-------
Lesson 3
Gas-Phase Contacting Scrubbers
Lesson Goal and Objectives
Goal
To familiarize you with the operation, collection efficiency, and maintenance problems of gas-
phase contacting scrubbers.
Objectives
Upon completing this lesson, you should be able to —
1. list three gas-phase contacting scrubbers and briefly describe how each operates,
2. recall operating characteristics such as pressure drop, liquid-to-gas ratio, and collection
efficiency (for both particulate and gaseous pollutants) of each of the above scrubbers,
and
3. describe typical operating and maintenance problems associated with each gas-phase
contacting scrubber design.
Introduction
Scrubbers using the exhaust (gas) stream to provide the energy for gas-liquid contact are
called gas-phase contacting scrubbers. The exhaust stream moves across or through a liquid
surface, shearing it to form tiny droplets. Breaking the liquid into fine droplets helps increase
both particle and gas collection. The droplets provide targets on which the particles hit and
are collected. They also provide a huge surface area for collecting (absorbing) gaseous
pollutants.
A number of methods are used to provide this shearing action. The gas can be forced
through cascades of liquid falling over flat plates. Holes can be punched in the plates, and the
gas can aspirate the water flowing over the plate. Or the gas can be forced through con-
stricted passages wetted with liquid, such as in orifice and venturi scrubbers. Three collectors
work primarily by this action: venturi scrubbers, plate towers, and orifice scrubbers.
3-1
-------
Venturi Scrubbers
A venturi scrubber is designed to effectively use the energy from the exhaust stream to atomize
the scrubbing liquid. Venturi devices have been used for over 100 years to measure fluid flow
(venturi tubes derived their name from G. B. Venturi, an Italian physicist). About 35 years
ago, Johnstone (1949) and other researchers found that they could effectively use the venturi
configuration to remove panicles from an exhaust stream. Figure 3-1 illustrates the classic
venturi configuration.
A venturi scrubber consists of three sections —a converging section, a throat section, and a
diverging section. The exhaust stream enters the converging section and, as the area
decreases, gas velocity increases. Liquid is introduced either at the throat or at the entrance to
the converging section. The exhaust gas, forced to move at extremely high velocities in the
small throat section, shears the liquid from its walls, producing an enormous number of very
tiny droplets. Particle and gas removal occur in the throat section as the exhaust stream mixes
with the fog of tiny liquid droplets. The exhaust stream then exits through the diverging sec-
tion, where it is forced to slow down. Venturis can be used to collect both paniculate and
gaseous pollutants, but they are more effective in removing panicles than in removing gaseous
pollutants.
Liquid can be injected at the converging section or at the throat. Figure 3-2 shows liquid
injected at the converging section. Thus, the liquid coats the venturi throat. This venturi is
very effective for handling hot, dry exhaust gas that contains dust. The dust would have a
tendency to cake on or abrade a dry throat. These Venturis are sometimes referred to as hav-
ing a wetted approach.
- Throat
Converging
section
Diverging
section
Figure 3-1. Venturi configuration.
3-2
-------
Figure 3-3 shows liquid injected at the venturi throat. Since it is sprayed at or just before
the throat, it does not actually coat the throat surface. These throats are susceptible to solids
buildup when the throat is dry. They are also susceptible to abrasion by dust panicles. These
Venturis are best used when the exhaust stream is cool and moist. In this venturi, the relative
particle-to-liquid velocity is the highest of any of the Venturis; therefore, the smallest particles
can be collected efficiently. These Venturis are referred to as having a non-wetted approach.
Throat
Figure 3-2. Venturi scrubber with a wetted throat.
Throat
Figure 3-3. Venturi with throat sprays.
3-3
-------
Venturis with round throats (Figures 3-2 and 3-3) can handle exhaust flows as large as
88,000 m3/h (40,000 cfm) (Brady and Legatshi 1977). At exhaust flow rates greater than this,
achieving uniform liquid distribution is difficult, unless additional weirs or baffles are used.
To handle large exhaust flows, scrubbers designed with long, narrow, rectangular throats
(Figure 3-4) have been used.
Liquid inlet
Figure 3-4. Spray venturi with rectangular throat.
Manufacturers have developed other modifications to the basic venturi design to maintain
scrubber efficiency by changing the pressure drop for varying exhaust gas rates. Certain types
of orifices that create more turbulence than a true venturi were found to be equally efficient
for a given unit of energy consumed (Mcllvaine Company 1974). Results of these findings led
to the development of the annular-orifice, or adjustable-throat, venturi scrubber (Figure 3-5).
The throat area is varied by moving a plunger, or adjustable disk, up or down in the throat,
decreasing or increasing the annular opening. Gas flows through the annular opening and
atomizes liquid that is sprayed onto the plunger or swirled in from the top.
3-4
-------
Another adjustable-throat venturi is shown in Figure 3-6. In this scrubber, the throat area is
varied by using a movable plate. A water-wash spray is used to continually wash collected
material from the plate.
Throat
Figure 3-5. Adjustable-throat venturi with plunger.
Movable plate
Figure 3-6. Adjustable-throat venturi with movable plate.
3-5
-------
Another modification can be seen in the venturi-rod scrubber. By placing a number of
pipes parallel to each other, a series of longitudinal venturi openings can be created as shown
in Figure 3-7. The area between adjacent rods is a small venturi throat. Water sprays help
prevent solids buildup. The principal atomization of the liquid occurs at the rods, where the
high-velocity gas moving through spacings creates the small droplets necessary for fine particle
collection.
All venturi scrubbers require an entrainment separator because the high velocity of gas
through the scrubber will have a tendency to exhaust the droplets. Cyclonic, mesh-pad, and
blade separators are all used. Cyclonic separators, the most popular, are connected to the ven-
turi vessel by a flooded elbow (Figure 3-8). The liquid reduces abrasion of the elbow as the
exhaust gas passes at high velocities from the venturi to the separator.
Figure 3-7. Venturi-rod scrubber.
Cyclonic
separator
Flooded
elbow
Figure 3-8. Flooded elbow leading into cyclonic separator.
3-6
-------
Particle Collection
Venturis are the most commonly used scrubber for particle collection and are capable, of
achieving the highest particle collection efficiency of any wet scrubbing system. As the exhaust
stream enters the throat, its velocity increases greatly, atomizing and turbulently mixing with
any liquid present. The atomized liquid provides an enormous number of tiny droplets for the
dust particles to impact on. These liquid droplets incorporating the panicles must then be
removed from the scrubber exhaust stream, generally by cyclonic separators.
Particle removal efficiency increases with increasing pressure drop (resulting in high gas
velocity and turbulence). Venturis can be operated with pressure drops ranging from 12 to
250 cm (5 to 100 in.) of water. Most Venturis normally operate with pressure drops in the
range of 50 to 150 cm (20 to 60 in.) of water. At these pressure drops, the gas velocity in the
throat section is usually between 30 and 120 m/s (100 and 400 ft/sec), or approximately
270 mph at the high end. These high pressure drops result in high operating costs.
The liquid-injection rate, or liquid-to-gas ratio (L/G), also affects panicle collection. The
liquid-injection rate depends on the temperature (evaporation losses) of the incoming exhaust
stream and the panicle concentration. Most venturi systems operate with an L/G ratio of 0.4
to 1.3 L/m3 (3 to 10 gal/1000 ft3) (Brady and Legatshi 1977). L/G ratios less than 0.4 L/m3
(3 gal/1000 ft3) are usually not sufficient to cover the throat, and adding more than 1.3 L/m3
(10 gal/1000 ft3) does not usually significantly improve particle collection efficiency.
Gay Collection
Venturi scrubbers can be used for removing gaseous pollutants; however, they are not used
when removal of gaseous pollutants is the only concern. The high exhaust gas velocities in a
venturi result in a very short contact time between the liquid and gas phases. This short con-
tact time limits gas absorption. However, venturi scrubbers are very useful for simultaneous
gaseous and particulate pollutant removal, especially when:
• scaling could be a problem,
• a high concentration of dust is in the exhaust stream,
• the dust is sticky or has a tendency to plug openings, and/or
• the gaseous contaminant is very soluble or chemically reactive with the liquid.
To maximize absorption of gases, Venturis operate at a set of conditions different from those
used to collect particles. Lower gas velocities and higher liquid-to-gas ratios are necessary for
efficient absorption. These values should be approximately 2.7 to 5.3 L/m3 (20 to 40
gal/1000 ft3). At high liquid-to-gas ratios, the gas velocity in the venturi throat is reduced (for
a given pressure drop). The reduction in gas velocity allows for a longer contact time between
phases.
3-7
-------
Maintenance Problems
The primary maintenance problem for venturi scrubbers is wear, or abrasion, of the scrubber
shell because of high gas velocities. Gas velocities in the throat can reach speeds of 430 km/h
(270 mph). Particles and liquid droplets traveling at these speeds can rapidly erode the scrub-
ber shell. Abrasion can be reduced by lining the throat with silicon carbide brick or fitting it
with a replaceable liner. Abrasion can also occur downstream of the throat section. To reduce
abrasion here, the elbow at the bottom of the scrubber (leading into the separator) can be
flooded. Particles and droplets impact on the pool of liquid, reducing wear on the scrubber
shell. Another technique to help reduce abrasion is to use a precleaner (i.e., quench sprays or
cyclone) to remove the larger particles.
The method of liquid injection at the venturi throat can also cause problems. Spray nozzles
are used for liquid distribution because they are more efficient (have a more effective spray
pattern) for liquid injection than are weirs. However, spray nozzles can easily plug when recir-
culating the liquid. Automatic or manual reamers can be used to correct this problem.
However, when heavy liquid slurries (either viscous or particle-loaded) are recirculated, open-
weir injection is often necessary. Table 3-1 summarizes some of the operational problems
associated with venturi scrubbers.
3-8
-------
Table 3-1. Operational problems associated with venturi scrubbers.
Problem
Probable cause
Corrective action
Low efficiency
Low scrubbing-liquid flow rate
Low pressure drop
Partially blocked entrainment
separator
Excessive gas flow
Inlet dust loading or particle size
distribution different from that
for which scrubber is designed
Check for plugged pipe or nozzles,
incorrectly opened valves, or over-
throttled pump-discharge valve.
Check for low scrubbing liquor, low gas
flow; inoperative or uncalibrated
variable-throat controller; damaged
variable-throat blade or disk.
Check washdown sprays, spray liquor
composition, and pH (for scaling).
Check for damper setting or venturi
throat setting.
High exit-gas
temperature
Low scrubbing-liquid rate
High water-inlet temperature
High inlet-gas temperature
Check for plugged pipe or nozzles,
incorrectly opened valves, or over-
throttled pump-discharge valve.
Check and adjust makeup or heat-
exchanger liquid flow rates.
Check quench sprays, if applicable, or
upstream equipment.
Exhaust gas liquor
entrainment
Plugged entrainment separator
Plugged moisture-eliminator drain
Excessive gas flow
Check washdown sprays and spray,
pattern; use more flushing periods if
necessary. Check liquor chemistry for
scaling agents.
Clean drain; add flushing water to con-
tinuously irrigate drain pipe.
Reduce gas flow.
Plugging or excessive
wear of spray
nozzles
Nozzle openings too small
Solids concentration too high in
spray liquor
Abrasives in spray liquor
Low pH in combination with
abrasives is causing erosion or
corrosion
Modify strainer/nozzle opening ratio so
that nozzle holes are at least twice the
diameter of strainer openings.
Check bleed line for malfunctions;
check for excessive dust loading. Check
strainers.
Remove abrasives or install abrasion-
resistant linings.
Check separation equipment. Check for
excessive dust loading in gas stream and
for purge-line malfunctioning. Remove
abrasives from liquor stream or install
abrasion-resistant linings in wear
zones. Add alkali for pH modification.
3-9
-------
Table 3-1. Operational problems associated with venturi scrubbers (continued).
Problem
Probable cause
Corrective action
Excessive throat wear
High solids recirculation
Excessive gas velocity
Corrosion/erosion
Check bleed line for malfunctions.
Check for excessive dust loading. Check
strainers.
Check throat pressure drop.
Check separation equipment. Check for
excessive dust loading in gas stream and
for purge-line malfunctioning. Add
alkali for pH modification. Install
abrasion-resistant liners in high-wear
zones if liquor modifications are not
practical.
Erratic automatic-
throat operation
Throat-mover malfunction
Sensor signal incorrect
Damaged damper-disk mechanism
Remove from service; repair or replace.
Check sensor taps for solids buildup.
Check transmission tubing for liquid
buildup or air leaks. Clean or repair
sensor.
First make external inspection of drive
train. If damaged area is not observed,
shut unit down and make internal
inspection using a throat-actuator man-
ual override. Check for packing damage
and excessively tight packing gland.
Low pressure drop
Broken, leaking, or plugged
static-tap line
Low gas-flow rate
Repair.
Check gas flow against design. Check
and, if necessary, adjust fan belt or
speed. Check inlet duct for obstructions.
Fan overloads and
shuts off
Excessive flow rate through fan
Low scrubbing-liquor flow rate
Check fan damper and variable-
throat opening.
Check for plugged pipe or nozzles,
incorrectly opened valves, or over-
throttled pump-discharge valve.
Wet-dry interface
buildup
Particle buildup where gas goes
from unsaturated to saturated
condition
Install special inlets. Reduce dissolved
solids in scrubbing liquor. Devote
routine maintenance to removal of
buildup.
Sources: Kelly 1978 and Anderson 2000 Co.
3-10
-------
Summary
Venturi scrubbers can have the highest particle collection efficiencies (especially for very small
particles) of any wet scrubbing system. They are the most widely used scrubbers because their
open construction enables them to remove most particles without plugging or scaling. Venturis
can also be used to absorb pollutant gases; however, they are not as efficient for this as are
packed or plate towers. The operating characteristics of venturi scrubbers are listed in
Table 3-2.
Table 3-2. Operating characteristics of venturi scrubbers.
Pollutant
Pressure drop
(Ap)
Liquid-to-gas
ratio
(L/G)
Liquid-inlet
pressure
(pi)
Removal efficiency
(%)
and cut diameter
for particles*
Gases
13-250 cm of water
(5-100 in. of water)
2.7-5.3 L/m5
(20-40 gal/1000 ft5)
<7-100 kPa
(<1-15 psig)
30-60% per venturi,
depending on
pollutant solubility
[90-99% is typical J
Particles
50-250 cm of water
["50-150 cm of water"]
|_ is common J
(20-100 in. of water)
[" 20-60 in. of water 1
|_ is common J
0.4-2.7 L/m5
(3.0-20.0 gal/1000 ft5)
0.2-fim cut diameter,
depending on Ap
*Cut diameter is the size (diameter) of the particle that is collected with at least 50% efficiency.
Venturi scrubbers have been designed to collect particles at very high collection efficiencies,
sometimes exceeding 99%. The ability of Venturis to handle large exhaust volumes at high
temperatures makes them very attractive to many industries; consequently, they have been
used to reduce particulate emissions in a number of industrial applications. This ability is par-
ticularly desirable for cement kiln emission reduction and for control of emissions from basic
oxygen furnaces in the steel industry, where the exhaust gas enters the scrubber at tempera-
tures greater than 350°C (660°F). Venturis have also been used to control fly ash and sulfur
dioxide emissions from industrial and utility boilers. A list of performance data for venturi
scrubbers is given in Table 3-3.
3-11
-------
Table 3-3. Performance data of typical venturi scrubbers.
Application
Dust
T ypical
particle-size
range
(/im)
Ap
(in. HtO)
Average
collection
efficiency
(%)
Iron and steel
Electric furnace
Ferro-manganese and
0.1-1.0
30-50
92-99
ferro silicon
BOF
Iron oxide
0.5-2.0
40-60
98.5
Gray iron cupola
Iron, coke, and silica
O
©
1
©
30-50
95
Mineral products
Asphalt dryer
Limestone and rock
1.0-50.0
10-15
98
Lime kiln
Lime
1.0-50.0
15-25
99
Cement kiln
Cement
0.5-55.0
10-15
97
Crushing and screening areas
Rock
0.5-100.0
6-20
99.9
Fertilizer manufacturing
Dryers
Ammonium chloride
0.05-1.0
10-20
85
fumes
Petroleum refining
Catalytic cracking unit
Catalyst dust
0.5-50.0
-
95
Chemical
Spray dryers
Fumes and odor
20-60
—
Phosphoric acid plant
Acid mist
40-80
98
Pulp and paper
Lime kiln
Soda ash
0.1 2.0
20-40
99
Recovery boiler
Salt particles
-
30-40
90
Boilers
Coal pulverizer
Fly ash
20 (mass median)
15-40
97-99
Stoker
Fly ash
75 (mass median)
10-15
97-99
3-12
-------
Review Exercise
1. Label the following sections of the venturi.
•S* •
'¦ *lH' '•*
- a.
-b.
- c.
2. In a venturi scrubber, the majority of pollutant
removal occurs in the
a. converging section.
b. throat.
c. diverging section.
1. a. converging
b. throat
c. diverging
3. A venturi scrubber in which liquid is introduced above
the throat section
a. increases the likelihood of dust buildup on the throat.,
b. reduces dust buildup on throat surfaces. ^
c. has the highest gas absorption capabilities of any wet
collector.
d. none of the above
2. b. throat
4. Many venturi scrubbers have devices by which the throat
area can be varied to maintain
a. gas velocity through the throat.
b. pressure drop. A
c. turbulence in the throat.
d. all of the above
3. b. reduces dust buildup
on throat surfaces.
5. True or False? Venturis are the most commonly used
scrubber for particle collection.
4. d. all of the above
5. True
3-13
-------
6. In a venturi scrubber, particle collection increases with
an increase in A O
7. Venturi scrubbers are generally limited in their capability
of removing gaseous pollutants because of
a. the short gas-liquid contact time.
b. low L/G ratios.
c. small liquid droplets formed in the throat.
d. all of the above
6. pressure drop
8. Venturi scrubbers are useful for simultaneous gas and
particle removal, especially when
a. scale buildup could be a problem.
b. a high concentration of dust is in the exhaust stream.
c. the dust is sticky or has a tendency to plug openings.
d. the gaseous pollutant is very soluble.
e. all of the above
7. a. the short gas-
liquid contact time.
9. To maximize gas collection in a venturi scrubber, the
pressure drop is increased/decreased and the L/ G Q
ratio is usually increased/decreased when compared ,
to operating conditions for particle collection.
8. e. all of the above
10. The primary maintenance problem for Venturis is
a. plugging due to the many internal pans. Q/
b. weeping due to low gas flows.
c. abrasion of the throat due to the high gas velocities.
d. all of the above
9. decreased,
increased
11. What does flooding the elbow between the venturi and the
separator reduce?
a. abrasion of the elbow
/ •
b. velocity of the gas stream p-/
c. plugging in the elbow
d. pressure drop across the device
10. c. abrasion of the
throat due to the
high gas velocities.
12. To be effective in collecting panicles, venturi scrubbers
must operate at a pressure drop of
a. 10 cm (5 in.) of water. \
b. 50 cm (20 in.) of water. :j
c. 150 cm (60 in.) of water.
d. any of the above
11. a. abrasion of the
elbow
13. In general, Venturis are more effective in removing
than rhev are in removing
a. gases, particles
b. particles, gases V
12. d. any of the above
(depending on the
specific scrubber
design and
application)
13. b. particles, gases
3-14
-------
Plate Towers
A plate tower is a vertical column with one or more plates (trays) mounted horizontally inside.
As shown in Figure 3-9, the exhaust stream enters at the bottom and flows upward, passing
through openings in the plates. Liquid enters at the top of the tower, traveling across each
plate to a downcomer through which it reaches either the next plate below or the bottom of
the tower. Pollutant collection occurs on each plate as the exhaust gas stream contacts and
then atomizes the liquid flowing over each plate. Plate towers are very effective in removing
gaseous pollutants and can be used simultaneously for particle removal. Plate towers may not
be appropriate when particle removal is the only consideration.
Underside
water sprays
Figure 3-9. Plate tower.
Plates, or trays, are designed in a variety of ways. The ones most commonly used for
industrial sources are the sieve, impingement, bubble-cap, and valve. The sieve, impingement,
and bubble-cap plates do not have moving parts, while the valve plates have liftable caps
above the opening in the plate. Plate openings can range from 0.32 to 2.50 cm (0.125 to
1.0 in.) in diameter for the sieve plate. Openings for the other plate designs are generally
larger.
3-15
-------
Sieve plates contain approximately 6456 to 32,280 holes per square meter (600 to 3000 per
square foot) of surface. Exhaust gas rises through these small holes and contacts the liquid at
the holes. The gas atomizes the liquid, forming a froth with droplets ranging from 10 to
100 fim in diameter. Particle collection efficiency increases as the size of the sieve opening
decreases. This is because of an increasing gas velocity and because smaller droplets are
formed. Sieve plates with large openings will not become plugged as easily as will other plate
designs. Figure 3-10 depicts gas-liquid contact on a sieve plate.
Impingement plates are similar to sieve plates with the addition of an impaction target
placed above each hole in the plate (Figure 3-11). The gas coming up through the hole forces
the liquid on the plate up against the target (impingement surface). This design increases the
mixing of the gas and liquid, provides an additional contact zone, and creates more liquid
droplets.
Figure 3-11. Impingement plate.
In the bubble-cap plate design, the exhaust gas enters each cap through a riser around
each hole in the plate and exits from several slots in each cap (Figure 3-12). This combination
of caps and risers creates a bubbly froth that allows good gas-liquid mixing, regardless of the
gas-to-liquid ratio. In addition, the caps provide a longer gas-liquid contact than either sieves
or impingement plates, thus increasing absorption efficiency. Plugging and corrosion can be a
problem for bubble-cap plates because of this more complex design.
In the valve plate design, the exhaust gas passes through small holes in the plate, pushing
up against a metal valve that covers each hole. The metal valve moves up and down with the
gas flow. The valve is limited in its vertical movement by legs attached to the plate (Figure
3-13). Therefore, the liftable valve acts as a variable orifice. Caps are available in different
weights to provide flexibility for varying exhaust gas flow rates. Floating valves increase
gaseous pollutant collection efficiency by providing adequate gas-liquid contact time, regard-
less of the exhaust gas flow rate. This design is also suited for very small particle collection;
however, valves will plug if large particles are in the exhaust stream. Wear and corrosion are
also a problem for the retaining legs. Valve plates are more expensive than sieve and impinge-
ment plates, but less expensive than bubble-cap plates.
3-16
-------
Figure 3-12. Bubble-cap plate.
Figure 3-13. Valve plate.
Particle Collection
Particles are collected in plate towers as the exhaust gas atomizes the liquid flowing over the
holes in the plates. The atomized droplets serve as impaction targets for the particles. Plate
towers are considered to be medium-energy scrubbers having moderate particle collection
efficiencies. Collection efficiency does not significantly increase by increasing the number of
plates over two or three. Collection efficiency can be increased by decreasing the hole size and
increasing the number of holes per plate. This produces more liquid droplets of a smaller size
and increases the gas velocity through the plate. However, it also increases the pressure drop
of the system.
Gas Collection
Plate towers are very effective for removing gaseous pollutants from an exhaust stream. They
can easily achieve greater than 98% removal in many applications. Absorption occurs as the
exhaust stream bubbles up through the liquid on the plates and contacts the atomized liquid
droplets. This action provides intimate contact between the exhaust gas and liquid streams,
allowing the liquid on each plate to absorb the pollutant gas. Each plate acts as a separate
absorption stage; therefore, absorption efficiency can be increased by adding plates. Absorp-
tion efficiency can also be improved by adding more liquid or by increasing the pressure drop
across each plate, which increases gas-liquid contact.
Maintenance Problems
Plate towers are susceptible to plugging and/or scale-buildup problems. If the exhaust stream
contains a high concentration of dust or sticky materials, plate towers are generally not used.
To clean the plates, access pons to each one are usually installed. In some systems, plates can
actually be removed for cleaning. In addition, water sprays can be used to spray the underside
of the lowest plate in the tower to eliminate the possibility of a wet-drv interface, which causes
plugging.
3-17
-------
Gas-liquid, distribution may also be a problem with plate towers. If the plates are not level,
gas-liquid contact will be reduced, thus reducing collection efficiency. Flooding (liquid
buildup on a plate) can occur if either the liquid-injection or exhaust gas velocity is excessive.
Flooding causes an increase in pressure drop and a decrease in gas-liquid mixing. Weeping
(liquid dripping through the holes in the plates) can occur if the gas velocity is too low. Weep-
ing also reduces gas-liquid contact. Table 3-4 summarizes some of the operational problems
associated with plate towers.
Table 3-4. Operational problems associated with plate towers.
Problem
Probable cause
Corrective action
Weeping
Too large an open area (holes) on
tray
Gas rate lower than design
Try bleeding in excess air or blocking
off excess area.
Check and adjust fan belt or speed.
Check inlet duct for obstructions.
Flooding
Too much liquid injected onto a
plate
Too much gas flowing through a
plate, causing the liquid to
"stand" on a plate
Reduce the liquid-injection rate.
Lower the gas flow rate, if possible.
If the gas flow rate is set because of
process conditions (and it is excessive),
an increase in tower design (size) may be
necessary.
Plugging
High solids concentration in
scrubbing liquor
Little or no water flow to trays
Higher-than-expected particle
content in inlet gas
Scale buildup
Check percentage of solids in recycle
liquid. Check solid-separation equip-
ment on recycle liquor. Use spray wash
header. Clean trays periodically.
Check pump output; look for plugged
piping, nozzles, incorrectly opened
valves, or overthrottled pump-
discharge valves.
Add prequench sprays.
Use a Iow-pH wash periodically to dis-
solve scale.
Poor distribution
Trays not level
Liquid flow rate too high or gas
flow rate too low
Mechanical problems with trays
Check and level.
Check pump output; look for plugged
piping, nozzles, incorrectly opened
valves, or overthrottled pump-discharge
valves.
Check for warped trays, loose fittings,
and loose or broken baffle strips or caps.
Sources: Kelly 1978 and Buonicore 1982.
3-18
-------
Summary
Plate towers are used most often when gaseous pollutant removal is the major concern. They
can achieve greater than 98% collection efficiency, depending on the solubility of the gaseous
pollutant. They can also be used to collect particles, but plugging and scale buildup problems
may occur. They have been used successfully in flue gas desulfurization systems to remove
sulfur dioxide emissions from utility boilers.
They have also been used to reduce pollutants emitted from petroleum refineries, chemical
processes, acid manufacturing plants, and metal smelters. A summary of the operating
characteristics of plate towers is given in Table 3-5.
Table 3-5. Operating characteristics of plate towers.
Pollutant
Pressure drop
(Ap)
Liquid-to-gas
ratio
(L/G)
Liquid-inlet
pressure
98%),
depending on
the solubility of
the gaseous
pollutant
Coal dryers
Copper roasting
Industrial boilers
Chemical process
industries
Petroleum
refineries
Incineration
processes
Panicles
Normal pressure
drops are 7.6 cm
(3 in.) of water
0.3-0.7 L/m3
(2-5 gal/1000 ft5)
>2.0-/im cut
diameter
3-19
-------
Review Exercise
1. Name each of the following designs for plates in plate
towers.
2. True or False? For particle collection, efficiency does not 1. a. bubble-cap
significantly increase by increasing the number of plates b. sieve
in a plate tower. , c. valve
3. In a plate tower, particle collection can usually be
enhanced by increasing/ decreasing the hole size
and/or/increasing/ decreasing the number of holes per
plate. v '
2. True
4. For gaseous pollutant collection in a plate tower, absorp-
3. decreasing,
tion can usually be enhanced by
increasing
a. adding plates.
b. increasing the amount of liquid.
c. both a and b
4. c. both a and b
3-20
-------
5. If rhe plares in a rower are nnr . Fas-liquid
contact can be reduced, thus reducing collection
efficiency.
a. the same size
b. level
c. staggered ^
d. omitted
6. Liquid dripping through the holes in the plates, due to a
low gas velocity, is referred to as
a. flooding.
b. dropsy. O
c. weeping.
d. drooling.
5. b. level
7. List at least three common operational problems
associated with plate towers.
6. c. weeping.
1
1
i
"--"D
7. • weeping
• Pegging
• flooding
3-21
-------
Orifice Scrubbers
In orifice scrubbers, the exhaust stream from the process is forced through a pool of liquid,
usually water. The exhaust stream moves through restricted passages, or orifices, to disperse
and atomize the water into droplets. These scrubbers are also called self-induced, spray, iner-
tial, or submerged orifice scrubbers.
Several orifice scrubber designs are typically used. In each, the incoming exhaust stream is
directed across or through a pool of water as shown in Figure 3-14. The high exhaust stream
velocity, approximately 15.2 m/s (50 ft/sec), creates a large number of liquid droplets. Both
particles and gaseous pollutants are collected as they are forced through the liquid pool and
impact the droplets. However, these scrubbers are generally used for removing particles. Large
particles are collected when they impact the liquid pool or its surface. Small particles are col-
lected when they impact the droplets. Baffles, or air foils, are added to provide turbulent mix-
ing of the exhaust stream and droplets.
Figure 3-14. Detail of orifice action.
In the self induced spray scrubber, the exhaust stream enters through a duct as shown in
Figure 3-15. The exhaust stream is forced by baffles through a pool of liquid. Particles and
gases are collected in the pool and by the droplets. Additional baffles placed in the path of
the "clean" exhaust stream as it exits the vessel serve as impingement surfaces to remove
entrained droplets.
Particulate matter collected in the scrubber forms a sludge that must be disposed of. Sludge
disposal involves removing and recycling large amounts of liquid, from 3.5 to 4.2 L/m3 (25 to
30 gal/1000 acfm). Some designs use a sludge separation and removal system inside the scrub-
ber. The water level inside the scrubber must be maintained during the sludge separation and
removal cycle so that the unit can operate efficiently.
3-22
-------
Figure 3-15. Self-induced spray orifice scrubber.
Particle Collection
Large particles in the incoming exhaust stream are collected as they impinge on the surface of
the pool. Smaller particles are collected as they impact on the droplets produced by the high-
velocity gas skimming over the liquid. Overall particle collection in an orifice scrubber
depends on the level of the liquid. The level of the liquid determines the gas velocity (and,
thus, the pressure drop) through the orifice. If the liquid level is low, gas velocities decrease
because the orifice opening is larger. Lower velocities produce fewer droplets that are larger in
size, decreasing particle collection. A turn-down of the system, or reduction in gas volume,
will also result in less atomization and produce larger droplets. It is recommended that gas
velocities should not fluctuate by more than 10 to 15% of design values to provide maximum
effectiveness (Bethea 1978).
Gas Collection
Orifice scrubbers are rarely used for absorption (Mcllvaine Company 1974). However, because
orifice scrubbers provide both thorough mixing of the gas and liquid, and large liquid-surface
contact areas (many tiny droplets), these devices can be effective for reactive scrubbing or for
removing gaseous pollutants that are already very soluble in the liquid. In reactive scrubbing,
the gaseous pollutants chemically react with the scrubbing liquid. These reactions occasionally
produce scale or sludge that can plug scrubber internals. The relatively large orifice openings
will not plug as easily as those in plate towers.
3-23
-------
Maintenance Problems
The greatest problem for orifice scrubbers is maintaining the liquid at the proper level for a
constant gas flow rate. Orifice scrubbers are designed to operate with a specific liquid level for
a given gas velocity. If the gas flow decreases (or the liquid level decreases), less atomization
occurs, thus reducing collection efficiency. If gas flow rate increases too much, it is possible to
blow the liquid chamber dry (Bethea 1978). Systems are generally designed to operate at the
upper end of the process exhaust rate and to introduce makeup air if the exhaust stream
velocity becomes too low. Controlling the liquid level is much more difficult than maintaining
a constant exhaust flow rate because of the turbulent condition of the water.
Summary
Orifice scrubbers are medium-energy devices with moderate collection efficiencies. The
pressure drops across these devices are usually between 5 and 25 cm (2 and 10 in.) of water.
The relatively large openings enable them to accommodate exhaust streams with high concen-
trations of paniculate matter. Plugging by sticky or stringy material and scale buildup are not
major problems. Because the gas stream is forced through a pool of liquid to create liquid
droplets, spray nozzles are not necessary.
Orifice scrubbers are used mainly on metallurgical processes (crushing, screening, grinding,
etc.), where the particles generated are mostly above 1 fim in diameter. Removal efficiencies
depend on exhaust stream velocities. Reduction in exhaust stream velocities or liquid levels in
the device will cause a reduction in collection efficiency. Table 3-6 lists operating
characteristics of orifice scrubbers.
Table 3-6. Operating characteristics of orifice scrubbers.
Pollutant
Pressure drop
(Ap)
Liquid-to-gas
ratio
(L/G)
Liquid-inlet
pressure
(pi)
Removal
efficiency
(%)
and cut
diameter
Applications
Gases
5-25 cm of water
(2- 10 in. of water)
0.07-0.7 L/m3
(0.5-5 gal/1000 ft3)
Not applicable
(nozzles are not
used)
Limited to very
soluble gases or
reactive
scrubbing
Mining operations
Rock products
industries
Foundries
Pulp and paper
industries
Chemical process
industries
Particles
1.3-5.3 L/m3
(10-40 gal/1000 ft3)
for sludge removal
and recycle
0.8-1-^m cut
diameter
3-24
-------
Review Exercise
1. Although orifice scrubbers are produced in a variety of
configurations, all are designed so that the exhaust gas
stream
a. is split into two streams.
b. travels concurrently with the liquid stream.
c. breaks through a pool of liquid. ^
d. none of the above
2. True or False? In an orifice scrubber, all particles are ^
collected as they impinge on the surface of the liquid. ^
1. c. breaks through a pool
of liquid.
3. The exhaust gas velocity (thus, the pressure drop) in an
orifice scrubber is dictated by the
a. adjustable throat.
b. level of liquid. ^
c. the plant foreman. ^
d. precise calculations.
2. False.
Only large particles.
Small particles are col-
lected by liquid droplets
produced by the exhaust
stream.
4. True or False? Orifice scrubbers are not primarily used
for gas absorption.
3. b. level of liquid.
5. The greatest problem with orifice scrubbers is
a. maintaining the proper liquid level.
b. plugging. f)
c. scale buildup. &
d. erosion.
4. True
6. True or False? An orifice scrubber is capable of operating
over a wide range of gas flow rates.
5. a. maintaining the
proper liquid level.
7. In orifice scrubbers, a reduction in the design exhaust gas
flow rate results in
a. an increase in gas collection.
b. an increase in particle collection.
c. less atomization and production of larger liquid
droplets.
d. both a and b
6. False.
Exhaust gas velocities
should not fluctuate
greatly.
8. Orifice scrubbers are ^enerallv classified as
energy devices capable of cnllerrinn
efficiencies. j
a. very low, very high ^
b. medium-, moderate J
i • i , • , IvXAju
c. high-, very high
d. none of the above
7. c. less atomization and
production of larger
liquid droplets.
8. b. medium-, moderate
3-25
-------
9. True or False? Plugging and scale buildup are not major
operating problems with orifice scrubbers. ^
9. True
References
Andersson 2000 Co. Venturi scrubbing equipment. Engineering manual with operating and
maintenance instructions. Atlanta, GA.
Beachler, D. S., andjahnke, J. A. October 1981. Control of particulate emissions. APTI
Course 413, EPA 450/2-81-066. U.S. Environmental Protection Agency.
Bethea, R. M. 1978. Air pollution control technology. New York: Van Nostrand Reinhold Co.
Brady, J. D., and Legatshi, L. K. 1977. Venturi scrubbers. In Air pollution control and
design handbook part 2. P. N. Chereminisoff and R. A. Young, eds. New York: Marcel
Dekker, Inc.
Buonicore, A.J. 1982. Wet scrubbers. In Air pollution control equipment, design, selection,
operation and maintenance. L. Theodore and A. J. Buonicore, eds. Englewood Cliffs:
Prentice-Hall, Inc.
Calvert, S. 1977. How to choose a paniculate scrubber. Chem. Eng. 84:133-140.
Johnstone, H. F., and Roberts, M. H. 1949. Deposition of aerosol particles from moving gas
streams. Ind. and Eng. Chem. 41:2417-2423.
Joseph, G. T., and Beachler, D. S. December 1981. Control of gaseous emissions. APTI
Course 415, EPA 450/2-81-005. U.S. Environmental Protection Agency.
Kelly, J. W. December 4, 1978. Maintaining venturi-tray scrubbers. Chem. Eng.
Mcllvaine Company. 1974. The wet scrubber handbook. Northbrook, IL.
Strauss, W. 1975. Industrial gas cleaning. Oxford: Pergamon Press, Inc.
3-26
-------
Lesson 4
Liquid-Phase Contacting Scrubbers
Lesson Goal and Objectives
Goal
To familiarize you with the operation, collection efficiency, and major maintenance problems
of liquid-phase contacting scrubbers.
Objectives
Upon completing this lesson, you should be able to—
1. list two liquid-phase contacting scrubbers and briefly describe the operation of each,
2. recall the operating characteristics such as pressure drop, liquid-to-gas ratio, and collec-
tion efficiency of each liquid-phase contacting scrubber, and
3. describe typical operating and maintenance problems associated with each design of
liquid-phase contacting scrubbers.
Introduction
The previous lesson described scrubbers that use the process gas stream as energy to atomize
liquid into collection droplets. Energy can also be applied to a scrubbing system by injecting
liquid at high pressure through specially designed nozzles. Nozzles produce droplets that fan
out into a spray in the scrubber chamber. Droplets act as targets for collecting particles
and/or absorbing gas in a pollutant exhaust stream. In liquid-phase contacting scrubbers, the
liquid-inlet pressure provides the major portion of the energy required for contacting the gas
(exhaust stream) and liquid phases.
Two liquid-phase contacting scrubbers are the spray tower and the ejector venturi. Many
other scrubber designs also incorporate sprays produced by nozzles, but in those scrubbers, the
sprays are used to clean trays or to wet scrubber surfaces and orifices, and not to provide the
gas-liquid contact in the system.
Spray Towers
Spray towers, or chambers, are constructed very simply — consisting of empty cylindrical vessels
made of steel or plastic and nozzles that are used to spray liquid in the vessels. The exhaust
stream usually enters the bottom of the tower and moves upward, while liquid is sprayed
4-1
-------
downward from one or more levels. This flow of exhaust gas and liquid in opposite directions
is called counter current flow. Figure 4-1 shows a typical countercurrent-flow spray tower.
Countercurrent flow exposes the exhaust gas with the lowest pollutant concentration to the
freshest scrubbing liquid.
Many nozzles are placed across the tower at different heights to spray all of the exhaust gas
as it moves up through the tower. The major purpose of using many nozzles is to form a
tremendous amount of fine droplets for impacting particles and to provide a large surface
area for absorbing gas. Theoretically, the smaller the droplets formed, the higher the collec-
tion efficiency achieved for both gaseous and particulate pollutants. However, the liquid
droplets must be large enough to not be carried out of the scrubber by the exhaust stream.
Therefore, spray towers use nozzles to produce droplets that are usually 500 to 1000 fim in
diameter. The exhaust gas velocity is kept low, from 0.3 to 1.2 m/s (1 to 4 ft/sec) to prevent
excess droplets from being carried out of the tower. Because of this low exhaust velocity, spray
towers must be larger than other scrubbers that handle similar exhaust stream flow rates.
Another problem occurring in spray towers is that after the droplets fall short distances, they
tend to agglomerate or hit the walls of the tower. Consequently, the total liquid surface area
for contact is reduced, thus reducing the collection efficiency of the scrubber.
In addition to a countercurrent-flow configuration, the flow in spray towers can be either a
cocurrent or crosscurrent configuration. In cocurrent-flow spray towers, the exhaust gas and
liquid flow in the same direction. Because the exhaust gas stream does not "push" against the
Liquid
sprays
Figure 4-1. Countercurrent-flow spray tower,
4-2
-------
liquid sprays, these units operate at higher exhaust gas velocities (through the vessels) than do
countercurrent-flow spray towers. Consequently, cocurrent-flow spray towers are smaller than
are countercurrent-flow spray towers (treating the same amount of exhaust flow).
In crosscurrent-flow spray towers, called horizontal-spray scrubbers, the exhaust gas and
liquid flow in directions perpendicular to each other (Figure 4-2). In this vessel, the exhaust
gas flows horizontally through a number of spray sections. The amount and quality of liquid
sprayed in each section can be varied, usually with the cleanest liquid (if recycled liquid is
used) sprayed in the last set of sprays.
Particle Collection
Spray towers are low-energy scrubbers. Contacting power is much lower than in venturi scrub-
bers, and the pressure drops across such systems are generally less than 2.5 cm (1 in.) of
water. The collection efficiency for small particles is correspondingly lower than in more
energy-intensive devices. They are adequate for the collection of coarse particles larger than
10 to 25 fim in diameter, although with increased liquid inlet nozzle pressures, particles with
diameters of 2.0 /an can be collected. Smaller droplets can be formed by higher liquid
pressures at the nozzle. The highest collection efficiencies are achieved when small droplets are
produced and the difference between the velocity of the droplet and the velocity of the
upward-moving panicles is high. Small droplets, however, have small settling velocities, so
there is an optimum range of droplet sizes for scrubbers that work by this mechanism.
Stairmand (1956) found this range of droplet sizes to be between 500 and 1000 /an for gravity-
spray towers. The injection of water at very high pressures, 2070 to 3100 kPa (300 to 450 psi),
creates a fog of very fine droplets. Higher particle-collection efficiencies can be achieved in
Liquid sprays
Blade entrainment
separator
Liquid-return
pumps
Figure 4-2. Crosscurrent-flow spray tower.
-------
such cases since collection mechanisms other than inertial impaction occur (Bethea 1978).
However, these spray nozzles may use more power to form droplets than would a venturi
operating at the same collection efficiency.
Gas Collection
Spray towers can be used for gas absorption, but they are not as effective as packed or plate
towers. Spray towers can be very effective in removing pollutants if the pollutants are highly
soluble or if a chemical reagent is added to the liquid. For example, spray towers are used to
remove HC1 gas from the tail-gas exhaust in manufacturing hydrochloric acid. In the produc-
tion of superphosphate used in manufacturing fertilizer, SiF4 and HF gases are vented from
various points in the processes. Spray towers have been used to remove these highly soluble
compounds. Spray towers are also used for odor removal in bone meal and tallow manufac-
turing industries by scrubbing the exhaust gases with a solution of KMnO*. Because of their
ability to handle large exhaust gas volumes in corrosive atmospheres, spray towers are also
used in a number of flue gas desulfurization systems as the first or second stage in the pollu-
tant removal process.
In a spray tower, absorption can be increased by decreasing the size of the liquid droplets
and/or increasing the liquid-to-gas ratio (L/G). However, to accomplish either of these, an
increase in both power consumed and operating cost is required. In addition, the physical size
of the spray tower will limit the amount of liquid and the size of droplets that can be used.
Maintenance Problems
The main advantage of spray towers over other scrubbers is that they are completely open;
they have no internal parts except for the spray nozzles. This feature eliminates many of the
scale buildup and plugging problems associated with other scrubbers. The primary
maintenance problems are spray-nozzle plugging or eroding, especially when using recycled
scrubber liquid. To reduce these problems, a settling or filtration system is used to remove
abrasive particles from the recycled scrubbing liquid before pumping it back into the nozzles.
(See Lesson 2 for additional information on spray nozzles.)
Summary
Spray towers are inexpensive control devices primarily used for gas conditioning (cooling or
humidifying) or for first-stage particle or gas removal. They are also being used in many flue
gas desulfurization systems to reduce plugging and scale buildup by pollutants. Many scrub-
bing systems use sprays either prior to or in the bottom of the primary scrubber to remove
large particles that could plug it. Spray towers have been used effectively to remove large par-
ticles and highly soluble gases. The pressure drops across the towers are very low (usually less
than 2.5 cm [1.0 in.] of water); thus, the scrubber operating costs are relatively low. However,
the liquid pumping costs can be very high.
Spray towers are constructed in various sizes —small ones to handle small exhaust flows of
0.05 m3/s (100 cfm) or less, and large ones to handle large exhaust flows of 50 m3/s
(100,000 cfm) or greater. Because of the low gas velocity required, units handling large
exhaust flow rates tend to be large in size. Operating characteristics of sprav towers are
presented in Table 4-1.
4-4
-------
Table 4-1. Operating characteristics of spray towers.
Pollutant
Pressure drop
(Ap)
Liquid-to-gas
ratio
(L/G)
Liquid-inlet
pressure
(pi)
Removal
efficiency
(%)
and cut
diameter
Applications
Gases
1.3-7.6 cm of water
(0.5-3.0 in. of water)
0.07-2.70 L/m3
(0.5-20 gal/1000 ft3)
(5 gal/1000 ft3 is
normal; >10 when
using pressure sprays)
70-2800 kPa
(10-400 psig)
50-90*%
(high efficiency
only when the
gas is very
soluble)
Mining industry
Chemical process
industry
Boilers and
incinerators
Iron and steel
industry
Particles
2-8-fj.m cut
diameter
Review Exercise
1. In a srnihher. the liquid and exhaust ?as flow
in opposite directions. (
a. cocurrent j/y
b. countercurrent
c. crosscurrent
d. Crosshatch
2. Tn a sprav rowpr the the droplet is. the
higher the theoretical collection efficiency will be.
a. smaller
b. larger
c. higher
d. lower
1. b. countercurrent
3. Gas velocities in spray towers are usually kept very
rn prevent excessive liquid from hernmintr
entrained in the exhaust gas stream leaving the tower.
a. high ^
b. low
c. stable
d. none of the above
2. a. smaller
4. True or False? In general, countercurrent-flow spray
towers must be larger than crosscurrent- or cocurrent-flow
spray towers to accommodate the same volumetric flow
rate.
3. b. low
4. True
4-5
-------
5. In a spray tower, gas collection can be increased by
increasing
a. the size of the liquid droplets.
b. the liquid-to-gas ratio (L/G). /
c. the gas velocity.
d. all of the above
6. Because spray towers contain few internal parts, they
a. eliminate many potential problems due to plugging and
scale buildup.
b. have low pressure drops.
c. are relatively simple and inexpensive. (y
d. all of the above
5. b. the liquid-to-gas ratio
(L/C?
7. What are the main maintenance problems with spray
towers?
6. d. all of the above
8. In spray towers, the pressure drops across the tower are "
usually lowV-high and the liquid pumping costs can be
very low7hig)t .
i 7. plugging or
erosion of the nozzle
8. low,
high
Ejector Venturis
The ejector, or jet, venturi scrubber uses a preformed spray, as does the simple spray tower.
The difference is that only a single nozzle is used instead of many nozzles. This nozzle operates
at higher pressures and higher injection rates than those in most spray chambers. The high-
pressure spray nozzle (up to 689 kPa or 100 psig) is aimed at the throat section of a venturi
constriction. Figure 4-3 illustrates the ejector venturi design.
The ejector venturi is unique among available scrubbing systems since it can move the proc-
ess gas without the aid of a blower or fan. The liquid spray coming from the nozzle creates a
partial vacuum in the side duct of the scrubber. This has the same effect as the water
aspirator used in high school chemistry labs to pull a small vacuum for filtering precipitated
materials (this is all due to the Bernoulli effect). This partial vacuum can be used to move the
process gas through the control device as well as through the process system. In the case of
explosive or extremely corrosive atmospheres, the elimination of a fan in the system can avoid
many potential problems.
The energy for the formation of scrubbing droplets comes from the injected liquid. The
high-pressure sprays passing through the venturi throat form numerous fine liquid droplets
that provide turbulent mixing between the gas and liquid phases. Very high liquid-injection
rates are used to provide the gas-moving capability and higher collection efficiencies. As with
other types of Venturis, a means of separating entrained liquid from the gas stream must be
installed. A liquid sump directs the gas flow to continuing ductwork. Entrainment separators
are commonly used to remove remaining small droplets.
4-6
-------
Figure 4-3. Ejector venturi scrubber.
Particle Collection
Ejector Venturis are effective in removing particles larger than 1.0 /on in diameter. These
scrubbers are not used on submicron-sized particles unless the panicles are condensable
(Gilbert 1977). Particle collection occurs primarily by impaction as the exhaust gas (from the
process) passes through the spray.
The turbulence that occurs in the throat area also causes the particles to contact the wet
droplets and be collected. Particle collection efficiency increases with an increase in nozzle
pressure and/or an increase in the liquid-to-gas ratio. In fact, ejector Venturis operate at
higher L/G ratios than most other particle scrubbers.
Gas Collection
Ejector Venturis have a short gas-liquid contact time because the exhaust gas velocities through
the vessel are very high. This short contact time limits the absorption efficiency of the system.
Although ejector Venturis are not used primarily for gas removal, they can be effective if the
gas is very soluble or if a very reactive scrubbing reagent is used. In these instances, removal
efficiencies of as high as 95% can be achieved (Gilbert 1977).
4-7
-------
Maintenance Problems
Ejector Venturis are subject to abrasion problems in the high-velocity areas —nozzle and
throat. Both must be constructed of wear-resistant materials because of the high liquid-
injection rates and nozzle pressures. Maintaining the pump that recirculates liquid is also very
important. In addition, the high gas velocities necessitate the use of entrainment separators to
prevent excessive liquid carryover. The separators should be easily accessible or removable so
that they can be cleaned if plugging occurs.
Summary
Because of their open design and the fact that they do not require a fan, ejector Venturis are
capable of handling a wide range of corrosive and/or sticky particles. However, they are not
very effective in removing submicron particles. They have an advantage in being able to
handle small, medium, and large exhaust flows. They can be used singly or in multiple stages
of two or more in series, depending on the specific application. Multiple-stage systems have
been used where extremely high collection efficiency of particles or gaseous pollutants was
necessary. Multiple-stage systems provide increased gas-liquid contact time, thus increasing
absorption efficiency. Table 4-2 lists the operating parameters for ejector Venturis.
Table 4-2. Operating characteristics of ejector Venturis.
Pollutant
Pressure drop
(Ap)
Liquid-to-gas
ratio
(L/G)
Liquid-inlet
pressure
(pt)
Removal
efficiency
(%)
and cut
diameter
Applications
Gases
1.3-13 cm of water
(0.5-5 in. of water)
7-13 L/m3
(50-100 gal/1000 ftJ)
100-830 kPa
(15-120 psig)
95% for very
soluble gases
Pulp and paper
industry
Chemical process
industry
Food industry
Metals- processing
industry
Particles
l-/un cut
diameter
4-8
-------
Review Exercise
1. The ejector, nr jer. venturi scrubber uses to
move the process exhaust stream.
a. multiple nozzles
b. a single high-pressure nozzle
c. a compressor
d. a fan
2. For ejector Venturis, particle collection efficiencies increase
with an increase in
a. nozzle pressure.
b. liquid-to-gas ratio (L/G).
c. pressure drop. ^
d. all of the above
1. b. a single high-pressure
nozzle
3. What limits gas collection in ejector Venturis? j
2. d. all of the above
4. Ejector Venturis are subject to abrasion problems in the ^
a. throat.
b. nozzle.
c. packing area.
d. throat and nozzle.
3. high gas velocities
5. True or False? Because of their open design and the fact
that they do not require a fan, ejector Venturis are capable
of handling a wide range of corrosive and/or sticky
particles. ^
4. d. throat and nozzle.
5. True
References
Beachler, D. S., and Jahnke, J. A. October 1981. Control of particulate emissions. APTI
Course 413, EPA 450/2-81-066. U.S. Environmental Protection Agency.
Bethea, R. M. 1978. Air pollution control technology. New York: Van Nostrand Reinhold
Co.
Gilbert, J. W. 1977. Jet venturi fume scrubbing. In Air pollution control and design hand-
book part 2. P. N. Cheremisinoff and R. A. Young, eds. New York: Marcel Dekker, Inc.
Joseph, G. T., and Beachler, D. S. December 1981. Control of gaseous emissions. APTI
Course 415, EPA 450/2-81-005. U.S. Environmental Protection Agency.
Mcllvaine Company. 1974. The wet scrubber handbook. Northbrook, IL.
4-9
-------
Lesson 5
Wet-Film Scrubbers
Lesson Goal and Objectives
Goal
To familiarize you with the operation, collection efficiency, and major maintenance problems
of wet-film scrubbers.
Objectives
Upon completing this lesson, you should be able to —
1. describe the operation of wet-film scrubbers,
2. recall the collection efficiency of wet-film scrubbers for particles and gases,
3. recognize at least three different gas-liquid flow arrangements (designs) for wet-film
scrubbers, and
4. recognize major operating and maintenance problems associated with each wet-film
scrubber design.
Introduction
In wet-film scrubbers, liquid is sprayed or poured over packing material contained between
support trays. A liquid film coats the packing through which the exhaust gas stream is forced.
Pollutants are collected as they pass through the packing, contacting the liquid film.
Therefore, both gas and liquid phases provide energy for the gas-liquid contact. These scrub-
bers are commonly called packed towers.
A wet-film scrubber uses packing to provide a large contact area between the gas and liquid
phases, to provide turbulent mixing of the phases, and to provide sufficient residence time for
the exhaust gas to contact the liquid. These conditions are ideal for gas absorption. Large
contact area and good mixing are also good for particle collection; however, once collected,
the particles tend to accumulate and, thus, plug the packing bed. The exhaust gas is forced to
make many changes in direction as it winds through the openings of the packed material.
Large particles unable to follow the streamlines hit the packing and are collected in the
liquid. As this liquid drains through the packing bed, the collected particles may accumulate,
thus plugging the void spaces in the packed bed. Therefore, wet-film scrubbers are not used
when particle removal is the only concern. Many other scrubber designs achieve better particle
removal for the same power input (operating costs).
5-1
-------
Gas Collection
For gas absorption, wet-film scrubbers are the most commonly used devices. The wet film
covering the packing enhances gas absorption several ways by providing:
• a large surface area for gas-liquid contact,
• turbulent contact (good mixing) between the two phases, and
• long residence time and repetitive contact.
Because of these features, packed towers are capable of achieving high removal efficiencies for
many different gaseous pollutants.
Numerous operating variables affect absorption efficiency. Of primary importance is the
solubility of the gaseous pollutants. Pollutants that are readily soluble in the scrubbing liquid
can be easily removed under a variety of operating conditions. Some other important
operating variables are discussed below.
Gas velocity—The rate of exhaust gas from the process determines the scrubber size to be
used. The scrubber should be designed so that the gas velocity through it will promote good
mixing between the gas and liquid phases. However, the velocity should not be too fast to
cause flooding.
Liquid-injection rate—Generally, removal efficiency is increased by an increase in the
liquid-injection rate to the vessel. The amount of liquid that can be injected is limited by
the dimensions of the scrubber. Increasing liquid-injection rates will also increase the
operating costs. The optimum amount of liquid injected is based on the exhaust gas flow
rate.
Packing size —Smaller packing sizes offer a large surface area, thus enhancing absorption.
However, smaller packing fits tighter, which decreases the open area between packing, thus
increasing the pressure drop across the packing bed.
Packing height —As packing height increases, total surface area and residence time
increase, enhancing absorption. However, more packing necessitates a larger absorption
system, which increases capital cost.
Tower Designs
Packed towers are typically designated by the flow arrangement used for gas-liquid contact or
by the material used as packing for the bed.
The most common flow configuration for packed towers is countercurrent flow. Figure 5-1
shows a packed tower with this arrangement. The exhaust stream being treated enters the bot-
tom of the tower and flows upward over the packing material. Liquid is introduced at the top
of the packing by sprays or weirs, and it flows downward over the packing material. As the
exhaust stream moves up through the packing, it is forced to make many winding changes in
direction, resulting in intimate mixing of both the exhaust gas and liquid streams. This
countercurrent-flow arrangement results in the highest theoretically achievable efficiency. The
most dilute gas is contacted with the purest absorbing liquor, providing a maximized concen-
tration difference (driving force) for the entire length of the column.
-------
Mist eliminator
Liquid sprays
Packing
Figure 5-1. Countercurrent-flow packed tower.
The countercurrent-flow packed tower does not operate effectively if there are large varia-
tions in the liquid or gas flow rates. If either the liquid-injection rate or the gas flow rate
through the packing bed is too high, a condition called flooding may occur. Flooding is a
condition where the liquid is "held" in the pockets, or void spaces, between the packing and
does not drain down through the packing. Flooding can be reduced by reducing the gas
velocity through the bed or by reducing the liquid-injection rate.
In another flow arrangement used with packed towers, cocurrent flow, both the exhaust
gas and liquid phases enter at the top-of the absorber and move downward over the packing
material. This allows the absorber to be operated at higher liquid and gas flow rates since
flooding is not a problem. The pressure drop is lower than with countercurrent flow since
both streams move in the same direction. The major disadvantage is that removal efficiency is
very limited due to the decreasing driving force (concentration differential) as the streams
travel down through the column. This limits the areas of application for cocurrent absorbers.
They are used almost exclusively in situations where limited equipment space is available,
since the tower diameter is smaller than a countercurrent or plate tower for equivalent flow
rates. Cocurrent flow is illustrated in Figure 5-2.
5-3
-------
Figure 5-2. Cocurrent-flow packed tower.
In packed towers using the cross flow arrangement, the exhaust gas stream moves horizon-
tally through the packed bed. The bed is irrigated by the scrubbing liquid flowing down
through the packing material. The liquid and exhaust gas flow in directions perpendicular to
each other. A typical crossflow packed tower is shown in Figure 5-3. (Inlet sprays aimed at the
face of the bed may also be included. If included, these sprays scrub both the entering gas
and the face of the packed bed.) The leading face of the packed bed is slanted in the direc-
tion of the oncoming gas stream. This ensures complete wetting of the packing by allowing
the liquid at the front face of the packing time to drop to the bottom before being pushed
back by the entering gas.
Crossflow absorbers are smaller and have a lower pressure drop than any other packed or
plate tower for the same application (removal efficiency and flow rates). In addition, they are
better suited than other wet-film scrubbers to handle exhaust streams with high particle con-
centrations. By adjusting the liquid flow rate, incoming particles can be removed and washed
away in the front half of the bed. This also results in a liquid savings by enabling the
crossflow packed tower to use less liquid in the rear sprays. This practice is carried one step
further by actually constructing the tower into sections as shown in Figure 5-4. The front sec-
tion can be equipped with water sprays and used for particulate matter removal. In the
second section, sprays may contain a reagent in the scrubbing liquor for gas removal. The last
section can be left dry to act as an entrainment separator. Crossflow packed towers do require
complex design procedures since concentration gradients exist in two directions in the liquid:
from top to bottom and from front to rear.
5-4
-------
/fliet sPrays L'^u'd
(°pti onaij
5Prays
re 5-3. Cjr<
PacAe^ «o»er.
ZJ,
quid
sprays
^%Ure
3"4-
Cr°ssflow,
Pacfced
fonder.
5-5
-------
Another crossflow packed tower is the fiber-bed, scrubber. The fiber-bed scrubber has
packed beds that are made with fibrous material such as fiberglass or plastic (Figure 5-5).
Liquid is sprayed onto the fiber beds to provide a wetted surface for pollutant removal and to
wash away any collected material.
Liquid sprays
Fiber bed
Figure 5-5. Fiber-bed scrubber.
Packing Material
Packing material is the heart of the tower. It provides the surface over which the scrubbing
liquid flows, presenting a large area for mass transfer to occur. Packing material represents
the largest material cost of the packed tower. Pictured in Figure 5-6 are some of the more
commonly used packings. These materials were originally made of stoneware, porcelain, or
metal, but presently a large majority are being made of high-density thermoplastics
(polyethylene and polypropylene). A specific packing is described by its trade name and
overall size. For example, a column can be packed with 5-cm (2-in.) Raschig rings® or 2.5-cm
(1-in.) Tellerettes® The overall dimensions of packing materials normally range from 0.6 to 10
cm (0.25 to 4 in.).
Specific packing selected for an industrial application depends on the nature of the con-
taminants, geometric mode of contact, size of the absorber, and scrubbing objectives. The
following factors provide a general guide for selecting packing materials (McDonald 1977):
Cost —Generally, plastic packing is less expensive than metal packing, with ceramic packing
being the most expensive. Packing costs are expressed in dollars per cubic meter ($/m3).
Low pressure drop—Presure drop is a function of the volume of void space in a cower
when filled with packing. Generally, the larger the packing size for a given bed size, the
smaller the pressure drop.
Corrosion resistance — Ceramic or porcelain packings are commonly used in a very corrosive
atmosphere.
5-6
-------
Berl saddle®
Raschig ring®
Intalox® Metal
Tellerette®
Figure 5-6. Common packing materials.
Large specific area —A large surface area per cubic foot of packing, m2/m3 (ft2/ft3), is
desirable for mass transfer.
Structural strength — Packing must be strong enough to withstand normal loads during
installation, service, physical handling, and thermal fluctuations. Ceramic packing may
crack under sudden temperature changes.
Weight — Heavier packing may require additional support materials or heavier tower con-
struction. Plastics have a big advantage in this area since they are much lighter than either
ceramic or metal packings.
Design flexibility — The efficiency of a scrubber changes as the liquid and gas flow rates
vary. Packing material must be able to handle the process changes without substantially
affecting removal efficiency.
Arrangement — Packing material may be arranged in an absorber in one of two ways. The
packing may be dumped into the column randomly or, in certain cases, systematically
stacked, as bricks are laid atop each other. Randomly packed towers provide a higher sur-
face area, m2/m3 (ft2/ft3), but also cause a higher pressure drop than stacked packing. In
addition to the lower pressure drop, the stacked packing provides better liquid distribution
over the entire surface of the packing. However, the large installation costs required to stack
che packing material usually make it impractical.
5-7
-------
Exhaust Gas Distribution
Uniform distribution of the exhaust gas through the packed beds is very important for effi-
cient pollutant removal. This is accomplished by properly designing the support trays used to
contain the packing in the bed. The support trays are essentially metal plates, or grids, that
support the packing while allowing the exhaust gas to flow evenly into the bed. If the packed
tower has multiple packing sections, each support grid would act as a distribution baffle,
directing the exhaust gas into the next packing section.
Liquid Distribution
As stated previously, one of the keys to effective packed tower operation is to intimately con-
tact the gas stream with the liquid stream. This contact must be maintained throughout the
entire column length. No packing material will adequately distribute liquid poured onto it at
only one point. Liquid introduced into the tower in this manner tends to flow down over a
relatively small cross section of the tower diameter. Known as liquid channeling, the liquid
flows in little streams down through the tower without wetting the entire packing area. Liquid
should be distributed over the entire cross-sectional top of the packing.
Once the liquid is distributed over the packing, it Rows down (by the force of gravity)
through the packing, following the path of least resistance. The liquid tends to flow toward
the tower wall, where the void spaces are greater than in the center. Once the liquid hits the
wall, it flows straight down the tower from that point (liquid channeling). A way must be pro-
vided to redirect the liquid from the tower wall back to the center of the column. This is
usually done by using liquid redistributors, which funnel the liquid back over the entire sur-
face of packing. It is recommended that redistributors be placed at intervals of no more than
3 m (10 ft) or 5 tower diameters, whichever is smaller (Zenz 1972).
Liquid can be distributed over the packing material by one of three devices: weirs, tubes, or
spray nozzles. Figure 5-7 shows both the commonly used weir and perforated-tube liquid
distributors. The drilled tube is often buried within the packing bed. This allows the liquid
coming out of the holes to be distributed over the packing without being blown against the
side walls of the tower. Burying the tube also allows the packing above the tube to act as an
entrainment separator.
Figure 5-7a. Trough and weir liquid distributor
Figure 5-7 b. Perforated-tube liquid distributor
5-8
-------
When packed towers are designed with spray nozzles, a few large nozzles will operate better
than will many small nozzles. Large nozzles are less susceptible to plugging. Small nozzles that
produce a finer spray are not needed in a packed tower because pollutant collection occurs on
the wetted packing and not by the liquid droplets. The advantages and disadvantages of each
liquid distributor are listed in Table 5-1.
Table 5-1. Liquid distributors for packed towers.
Distributor
Advantages
Disadvantages
Weirs
Handle dirty liquids with a high
solids content
Can use river or unfiltered water
Can be easily inspected and main-
tained if access is available
Most costly to purchase
Do not distribute liquid as uni-
formly as other methods
Weirs must be level
Tubes
Uniform liquid distribution
Can be buried below packing
surface
Generally least expensive to
purchase
Easily plugged, must use filter
Difficult to determine if holes are
plugged when tube is buried in
the packing
Spray nozzles
Uniform liquid distribution
Tower need not be plumb
Can be easily inspected and main-
tained if access is available
Highest pressure drops and opera-
ting costs
Easily plugged, must use filter
Source: Clark 1975.
Review Exercise
1. In a packed rower, rhe tras srream beinp-
treated enters the bottom and flows upward through the
packing while the liquid is introduced over the top of the
packing and flows down through it.
a. cocurrent
b. crossflow
c. countercurrent
2. A packed rower rannor handle larp-e varia-
tions in liquid or gas flow rates because flooding may
occur.
a. cocurrent
b. countercurrent
c. crossflow
d. fiber-bed
1. c. countercurrent
2. b. countercurrent
5-9
-------
3. Cocurrent packed towers usually have higher/lower /)
pressure drops than do countercurrent packed towers. A~
4. True or False? Crossflow packed towers can handle flue
gas containing a high concentration of particulate matter
because they use liquid sprays that will remove and wash /
away particles in the front half of the bed. /\
3. lower
5. Packing material is usually made of
a. porcelain.
b. polyethylene.
c. polypropylene. ^
d. all of the above
4. True
6. Packed towers that have been raji^Lomly packed have
more/less, surface area and highepr/lower pressure drops
th
-------
Another critical problem in packed tower operation is maintaining the proper liquid and
gas flow rates. If the liquid or gas flow rate increases (one in relation to the other), a point is
reached where the rising exhaust gas starts to hold up the descending liquid. The liquid fills
the upper portion of the packing until its weight causes it to fall. This condition, known as
flooding, results in a high pressure drop, a pulsating airflow in the tower, and greatly reduced
pollutant removal efficiencies. Optimum operating flow rates are normally at 60 to 75% of
the flooding conditions. Conversely, a gas flow rate that is too low can also cause mixing prob-
lems. Too low a gas velocity can result in gas channeling. Gas channeling occurs when the gas
does not distribute uniformly through the packing, but moves only through a small portion of
the bed (following the path of least resistance). This normally occurs near the walls of the
tower, where the void spaces are the greatest. Table 5-2 lists problems that could occur in
daily operation of a packed tower and some probable causes of these problems.
Table 5-2. Potential operating problems and causes for packed towers.
Problem
Possible causes
Static pressure drop
increases
Liquid flow rate to liquid distributor has
increased and should be checked.
Packing in irrigated bed could be partially
plugged due to solids deposition, and may
require cleaning.
Entrainment separator could be partially
plugged and may require cleaning.
Packing support plate at bottom of packed
section could be blinded, causing increased
pressure drop, which will require cleaning.
Packing could be settling due to corrosion or
solids deposition, again requiring cleaning
or additional packing.
Airflow rate through absorber could have
been increased by a change in damper set-
ting, which may need readjustment.
Pressure drop decreases,
slowly or rapidly
Liquid flow rate to distributor has decreased
and should be adjusted accordingly.
Airflow rate to scrubber has decreased due to
a change in fan characteristics or due to a
change in system damper settings.
Partial plugging of spray or liquid distribu-
tor, causing channeling through scrubber,
could be occurring. Liquid distributor should
be inspected to ensure that it is totally
operable.
Packing support plate could have been dam-
aged and fallen into bottom of the absorber,
allowing packing to fall to bottom and pro-
duce a lower pressure drop. This should be
checked.
5-11
-------
Table 5-2. Potential operating problems and causes for packed towers
(continued).
Problem
Possible causes
Pressure or flow change
in recycled liquid
causing reduced liquid
flow
Plugged strainer or filter in recycle piping,
which may require cleaning.
Plugged spray nozzles, which may require
cleaning.
Piping may be becoming partially plugged
with solids and need cleaning.
Liquid level in sump could have decreased,
causing pump cavitation.
Pump impeller could have been worn
excessively.
Valve in either suction or discharge side of
pump could have been inadvertently closed.
High liquid flow
Break in the internal distributor piping.
Spray nozzle that has been inadvertently
"uninstalled."
Spray nozzle that may have come loose or
eroded away, creating a low pressure drop.
Change in throttling valve setting on the dis-
charge side of the pump, allowing larger
liquid flow; reset to the proper conditions.
Excessive liquid
carryover
Partially plugged entrainment separator,
causing channeling and reentrainment of the
collected liquid droplets.
Airflow rate to absorber could have increased
above the design capability, causing
reentrainment.
If a packed-type entrainment separator was
used, packing may not be level, causing chan-
neling and reentrainment of moisture.
If a packed entrainment separator was used,
and a sudden surge of air through the
absorber occurred, this could have caused the
packing to be carried out of absorber or to be
blown aside, creating an open area "hole"
through separator.
Velocity through absorber has decreased to a
point that absorption does not effectively take
place, and low removal is achieved.
5-12
-------
Table 5-2. Potential operating problems and causes for packed towers
(continued).
Problem
Possible causes
Reading indicating low
airflow
Packing in absorber may be plugged, causing
a restriction to airflow.
Liquid flow rate to absorber could have been
increased inadvertently, again causing
greater restriction and pressure drop,
creating lower gas flow rate.
Fan belts have worn or loosened, reducing
airflow to equipment.
Fan impeller could be partially corroded,
reducing fan efficiency.
Ductwork to or from absorber could be par-
tially plugged with solids and may need
cleaning.
Damper in system has been inadvertently
closed or setting changed.
Break or leak in duct could have occurred due
to corrosion.
Increase in airflow
Sudden opening of damper in system.
Low liquid flow rate to absorber.
Packing has suddenly been damaged and has
fallen to bottom of absorber.
Sudden decrease in
absorber efficiency
Liquid makeup rate to the absorber has been
inadvertently shut off or throttled to a low
level, decreasing absorber efficiency.
If a chemical feed system is employed, system
may have run out of chemical required; this
could indicate malfunction of pH probes, if
employed, requiring replacement.
Set point on pH control may have to be
adjusted to allow more chemical feed.
Problem may exist with chemical metering
pump, control valve, or line pluggage.
Liquid flow rate to scrubber may be too low
for effective removal.
Source: MacDonald 1982.
5-13
-------
Summary
Packed towers are mainly used to remove gaseous pollutants. Because of plugging problems,
they are not used when particle removal is the only concern, or when a high concentration of
particulate matter is in the exhaust gas. Packed towers are capable of very high efficiencies for
removing many gaseous pollutants. For pollutants that are only slightly soluble, or if the
gaseous pollutant removal efficiency must be greater than 99%, packed towers or plate towers
can be used. Plate towers are used to control emissions from many of the same processes that
would use packed towers. These units were described in Lesson 3.
The following list gives some factors used in comparing plate towers to packed towers:
1. Packed towers are not able to handle particulate matter or other solid materials in the
flue gas as well as plate towers.
2. Plate towers are chosen for operations that require a large number of transfer units or
that must handle large gas volumes. To achieve the same collection efficiency (transfer
units), packed towers must have either deep packed beds or multiple beds. Packed
towers can experience liquid channeling problems if the diameter or height of the tower
is too large. Redistribution trays must be installed in large-diameter and tall packed
towers to avoid channeling.
3. The total weight of a packed tower is more than that of a comparable plate tower.
4. Packed towers are much cheaper to construct than plate towers if corrosive substances
are to be handled. Packed towers can be constructed with a fiberglass-reinforced
polyester shell which is generally about half the cost of a carbon steel plate tower.
5. Packed towers cannot handle volume and temperature fluctuations as well as plate
towers. Expansion or contraction due to temperature changes can crush or melt packing
material.
In a packed tower, the optimum pressure drop through a packing section is 1.7 to 5.0 cm
(0.2 to 0.6 in.) of water per foot of installed packing (Clark 1975). The overall pressure drops
across packed towers are usually between 5 and 25 cm (2 and 10 in.) of water. Thus, packed
towers are generally considered as medium-energy scrubbers.
Packed towers are most suited to applications where a high gas-removal efficiency is
required and the exhaust gas is relatively free from particles. These include removing HC1,
NH4, and S02 gases from a variety of process streams such as those from fertilizer manufac-
turing, chemical processing, acid manufacturing, steel making, and metal pickling operations.
One important point that should be noted is that packed towers are not effective in removing
submicron-sized particles, even if the particles are very soluble. Inorganic salts or fumes such
as ammonium chloride or aluminum chloride are prime examples. These particles are usually
so small that they flow with the exhaust gas through the packing bed and are not absorbed.
Table 5-3 lists the general operating characteristics of wet-film scrubbers.
5-14
-------
Table 5-3. Operating characteristics of wet-film scrubbers.
Pollutant
Pressure drop
(Ap)
Liquid-to-gas
ratio
(L/G)
Liquid-inlet
pressure
(p£)
Removal
efficiency
(%)
and cut
diameter
Applications
Gases
2-8.5 cm/m
of column packing
(0.25-1 in./ft
of column packing)
0.13-2.0 L/m5
(1-15 gal/1000 ft',
depending on type of
flow and packing)
34-100 kPa
(5-15 psig)
Very high,
99*%, depend-
ing on operating
conditions
Mainly used for
gaseous pollu-
tant removal
Metal operations
Acid plants
Chemical process
industries
Particles
1.5-/un cut
diameter
Review Exercise
1. Packed towers are frequently used for removing gaseous
pollutants because
a. the packing provides a large surface area for gas-liquid
contact.
b. they have relatively low pressure drops.
c. the packing provides good mixing of gas and liquid and
a long residence time. \
d. all of the above (^3>
2. Increasing the liquid flow rate in a packed tower will
usually increase/decrease the gas removal rate.
1. d. all of the above
3. If the gas flow rate through a packed tower is too low,
may occur.
a. flooding
b. mixing
c. gas channeling
d. plugging
2. increase
4. True or False? Packed towers remove particulate matter
and other solids more easily and with less maintenance
problems than do plate towers. Q-
3. c. gas channeling
4. False
5-15
-------
5. In a packed tower, liquid occasionally flows in little
streams straight through the packing without wetting the
packing surface. This condition is called
a. flooding.
b. liquid channeling. 1
c. mixing. Kj-
d. plugging.
6. In processes having high-temperature flue gas, (
plate/packed towers are more suitable because their
internal components will expand and contract. ^
5. b. liquid channeling
7. Packed towers are most suitable for industrial processes
requiring high gas-removal efficiency, but not having a
high concentration of particulate matter in the flue gas. In
which of the following industrial processes would a packed
tower be an appropriate control device?
a. coal-fired boiler
b. a pickling tank using HC1 (in the steel industry)
c. nitric acid plant
d. ammonium chloride production
e. both b and c
6. plate
7. e. both b and c
References
Beachler, D. S., andjahnke, J. A. October 1981. Control of particulate emissions. APTI
Course 413, EPA 450/2-81-066. U.S. Environmental Protection Agency.
Bethea, R. M. 1978. Air pollution control technology. New York: Van Nostrand Reinhold
Co.
Clark, J. M. 1975. Absorption equipment. In Handbook for the operation and maintenance
of air pollution control equipment. F. L. Cross and H. E. Hesketh, eds. Westport:
Technomic Publishing Co., Inc.
Joseph, G. T., and Beachler, D. S. December 1981. Control of gaseous emissions. APTI
Course 415, EPA 450/2-81-005. U.S. Environmental Protection Agency.
Mcllvaine Company. 1974. The wet scrubber handbook. Northbrook, IL.
MacDonald, J. W. 1982. Absorbers. In Air pollution control equipment, design, selection,
operation and maintenance. L. Theodore and A. J. Buonicore, eds. Englewood Cliffs:
Prentice-Hall, Inc.
MacDonald, J. W. 1977. Packed wet scrubbers. In Air pollution control and design handbook
part 2. P. N. Cheremisinoff and R. A. Young, eds. New York: Marcel Dekker, Inc.
Zenz, F. A. 1972. Designing gas absorption towers. Chem. Eng. 79:120-138.
5-16
-------
Lesson 6
Combination Devices—
Liquid-Phase and Gas-Phase
Contacting Scrubbers
Lesson Goal and Objectives
Goal
To familiarize you with the operation, collection efficiency, and major maintenance problems
of scrubbers that use energy from both the liquid and gas phases for contact.
Objectives
Upon completing this lesson, you should be able to —
1. list four combination contacting scrubbers,
2. describe the operation of each combination contacting scrubber listed above,
3. recall the relative collection efficiency of each scrubber for both panicles and gaseous
pollutants, and
4. name the major operating or maintenance problems associated with each device.
Introduction
A number of wet-collector designs use energy from both the gas stream and liquid stream to
collect pollutants. Many of these combination devices are available commercially. A seemingly
unending number of scrubber designs have been developed by changing system geometry and
incorporating vanes, nozzles, and baffles. This lesson will describe several systems that
incorporate features of both liquid-phase and gas-phase contacting wet collectors.
Cyclonic Spray Scrubbers
Cyclonic spray scrubbers use the features of both the dry cyclone and the spray chamber to
collect pollutants. Generally, the exhaust gas enters the chamber tangentially, swirls through
the chamber in a corkscrew motion, and exits. At the same time, liquid is sprayed inside the
chamber. As the exhaust gas swirls around the chamber, pollutants are captured when they
impact on liquid droplets, are thrown to the walls, and washed back down and out.
6-1
-------
Cyclonic scrubbers are generally low- to medium-energy devices, with pressure drops of 4 to
25 cm (1.5 to 10 in.) of water. Commercially available designs include the irrigated cyclone
scrubber and the cyclonic spray scrubber. In the irrigated cyclone (Figure 6-1), the exhaust
gas enters near the top of the scrubber into the water sprays. The exhaust gas is forced to
swirl downward, then change directions, and return upward in a tighter spiral. The liquid
droplets produced capture the pollutants, are eventually thrown to the side walls, and carried
out of the collector. The "cleaned" gas leaves through the top of the chamber.
The cyclonic spray scrubber (Figure 6-2) forces the exhaust gas up through the chamber
from a bottom tangential entry. Liquid sprayed from nozzles on a center post (manifold) is
directed toward the chamber walls and through the swirling exhaust gas. As in the irrigated
cyclone, liquid captures the pollutant, is forced to the walls, and washes out. The "cleaned"
gas continues upward, exiting through the straightening vanes at the top of the chamber.
Spray ring
Figure 6-1. Irrigated cyclone scrubber.
Straightening vanes
Spray manifold
1PJ"
T Water in
Water out f
Figure 6-2. Cyclonic spray scrubber.
6-2
Mm
-------
Stationary vanes are used inside the cyclonic scrubber chamber for much the same purpose
that they are used at the top —to alter the gas flow. But inside, they are designed to start or
enhance the cyclonic gas flow.
Particle Collection
Cyclonic spray scrubbers are more efficient than spray towers, but not as efficient as venturi
scrubbers, in removing particles from the exhaust gas stream. Particles larger than 5 /im are
generally collected by impaction with 90% efficiency. The cut diameter ranges between 2 and
3 fim for these devices. In a simple spray tower, the velocity of the particles in the exhaust gas
stream is low: 0.6 to 1.5 m/s (2 to 5 ft/sec). By introducing the exhaust gas tangentially into
the spray chamber, as does the cyclonic scrubber, exhaust gas velocities (thus, particle
velocities) are increased to approximately 60 to 180 m/s (200 to 600 ft/sec). The velocity of
the liquid spray is approximately the same in both devices. This increased particle-to-liquid
relative velocity increases particle collection efficiency for this device over that of the spray
chamber. Exhaust gas velocities of 60 to 180 m/s are equivalent to those encountered in a
venturi scrubber. However, cyclonic spray scrubbers are not as efficient as Venturis because
they are not capable of producing the same degree of useful turbulence.
Gas Collection
High exhaust gas velocities through these devices reduce the gas-liquid contact time, thus
reducing absorption efficiency. Cyclonic spray scrubbers are capable of effectively removing
some gases; however, they are rarely chosen when gaseous pollutant removal is the only
concern.
Maintenance Problems
The main maintenance problems with cyclonic scrubbers are nozzle plugging and corrosion or
erosion of the side walls of the cyclone body. Nozzles have a tendency to plug from particles
that are in the recycled liquid and/or panicles that are in the exhaust gas stream. The best
solution is to install the nozzles so that they are easily accessible for cleaning or removal. Due
to high gas velocities, erosion of the side walls of the cyclone can also be a problem. Abrasion-
resistant materials may be used to protect the cyclone body, especially at the inlet.
Summary
The pressure drops across cyclonic scrubbers are usually 4 to 25 cm (1.5 to 10 in.) of water;
therefore, they are low- to medium-energy devices and are most often used to control large-
sized panicles. Relatively simple devices, they resist plugging because of their open construc-
tion. They also have the additional advantage of acting as entrainment separators because of
their shape. Their biggest disadvantages are that they are not capable of removing submicron
particles and they do not efficiently absorb most pollutant gases. Table 6-1 lists typical
operating characteristics of cyclonic scrubbers.
6-3
-------
Table 6-1. Operating characteristics of cyclonic scrubbers.
Pollutant
Pressure drop
(Ap)
Liquid-to-gas
ratio
(L/G)
Liquid-inlet
pressure
(pi)
Removal
efficiency
<%>
and cut
diameter
Applications
Gases
4-25 cm of water
(1.5-10 in. of water)
0.3-1.3 L/m3
(2-10 gal/1000 ft3)
280-2800 kPa
(40-400 psig)
Only effective
for very soluble
gases
Mining operations
Drying operations
Food processing
Foundries
Particles
2-3-;Mn cut
diameter
Review Exercise
1. Cyclonic spray scrubbers are more efficient than
. hut not as efficient as . in remov-
ing particles.
a. spray towers, venturi scrubbers ,yc
b. venturi scrubbers, spray towers
2. In a cyclonic spray scrubber, panicles are primarily
collected
a. as they hit the wetted walls.
b. as they impact the liquid droplets. \ -
c. due to gravity.
d. in the throat.
1. a. spray towers,
venturi scrubbers
3. True or False? Cyclonic scrubbers are not often used to
control gaseous emissions.
2. b. as they impact the
liquid droplets.
4. The main maintenance problems associated with cyclonic
scrubbers are and
3. True
5. Cvclonic scrubbers are eneriry devires. '
a. high ^
b. low- to medium-
4. nozzle plugging (and)
corrosion or erosion of
the side walls in the
chamber
6. What are cyclonic scrubbers used most often to control?
a. micron-sized particles
b. large-sized particles
c. gaseous emissions
d. particles and gases simultaneously
5. b. low- to medium-
6. b. large-sized particles
6-4
-------
Mobile-Bed Scrubbers
Mobile-bed, also called moving-bed, scrubbers use energy from both liquid sprays and the gas
stream to provide contact. Instead of having stationary packing, as in packed towers, they use
a bed containing packing that is in constant motion. The gas stream provides the energy to
keep the packing in motion while, at the same time, liquid is sprayed over the packing.
Mobile-bed scrubbers can be classified as either flooded or fluidized, depending on the degree
of packing movement. In a flooded-bed scrubber, the packing gently moves and rotates,
whereas in a fluidized scrubber, the packing is suspended, or fluidized, within the bed.
Mobile-bed scrubbers were developed to provide the effective mass-transfer (absorption)
characteristics of packed and plate towers, without the plugging problems. The wetted pack-
ing provides a large area for gas-to-liquid contact, promoting absorption. Hie movement of
the bed cleans off any deposited particles. Therefore, these devices are primarily used when
good collection efficiency for both particulate and gaseous pollutants is required.
A flooded-bed scrubber (Figure 6-3) contains a section of mobile packing (spheres) 10 to 20
cm (4 to 8 in.) deep. The spheres are usually made of plastic; however, glass or marble
spheres have been used. The exhaust gas stream enters from the bottom while liquid is
sprayed from the top and/or bottom over the packing. Bottom, or inlet, sprays are usually
included to saturate the exhaust gas stream and remove any large particles. The gas velocity is
such that it causes the packing materials to rotate and rub against each other. This rotating
motion acts as a self-cleaning mechanism in addition to enhancing gas and liquid mixing.
Mobile packii
Mist eliminator
Spray bar
Figure 6-3. Flooded-bed scrubber.
6-5
-------
Bubbles formed in the bed create a layer of froth over the bed approximately twice as deep as
the bed itself. This turbulent froth layer provides an additional surface for absorbing pollutant
gases and collecting fine particles. Because of the high gas velocities, entrainment separators
are required to prevent liquid-mist carryover.
A fluidized-bed scrubber is very similar to a flooded-bed scrubber. The difference is in the
degree of movement of the packing. In a fluidized-bed scrubber, the exhaust gas velocity (1.8
to 4.8 m/s, or 6 to 16 ft/sec) is such that it keeps the packing in constant motion between a
lower and upper retaining grid. This is shown in Figure 6-4. The packing is made of either
polypropylene or polyethylene plastic balls that are hollow, resembling ping pong balls. The
packed sections are usually 0.3 to 0.6 m (1 to 2 ft) thick with a froth zone about 0.6 m (2 ft)
thick above the packing. These devices can have one to as many as six fluidized packed sec-
tions. When used for gas absorption, they are sometimes referred to as turbulent-contact
absorbers (TCA).
Mobile packing
Figure 6-4. Fluidized-bed scrubber.
6-6
-------
Particle Collection
In a mobile-bed scrubber, panicles can be collected in three locations. First, sprays are used
to remove coarse particles in the inlet below the bed. Particles are also captured when they
impinge on the wetted surface of the packing. Finally, small panicles are captured in the
froth, or foam, layer above the bed. These devices will generally remove panicles as small as 2
to 3 fxm in diameter and have been used extensively when the exhaust stream does not contain
a substantial amount of particles in the submicron range. These devices usually contain one
bed, unless gas absorption is a consideration. Adding additional beds or more packing does
not substantially increase the panicle collection efficiency (i.e., any panicles not captured by
the first stage will probably not be collected in any following stages). The pressure drop in
mobile-bed scrubbers ranges from 5 to 15 cm (2 to 6 in.) of water per stage of packing.
Gas Collection
Mobile-bed scrubbers are capable of achieving high gaseous-polluant removal efficiencies.
Their operation is very similar to the operation of packed towers. Liquid is sprayed over the
mobile packing, providing a huge surface for the pollutant gas to contact the liquid. Move-
ment of both the gas around the packing and the constantly sprayed liquid provides excellent
mixing and contact time for absorption to occur. Mobile-bed scrubbers provide the same
amount of absorption efficiency as do packed or plate towers without the associated plugging
problems. Due to the high exhaust gas velocities through mobile-bed scrubbers, these units
can handle five to six times the amount of exhaust gas handled by packed or plate towers of
similar size (Bethea 1978). However, they are not as efficient as packed or plate towers per
unit of energy consumed.
Absorption in mobile-bed scrubbers is enhanced by the same factors that affect packed
towers. These factors are increasing the liquid-to-gas ratio, increasing the depth of packing, or
increasing the number of stages. Increasing these factors increases the gas and liquid contact
and the residence time. However, increasing these factors also increases the capital and/or
operating costs of the system. As with any system, these process variables are set to achieve the
desired removal efficiency at the minimum cost. For gas absorption, multiple stages are used
and the liquid-to-gas ratios are high. For example, mobile-bed scrubbers have been used to
remove S02 from boiler flue gas exhausts. Using a lime or limestone slurry, the liquid-
injection rates are approximately 8 L/m3 (60 gal/1000 ft3) of flue gas. This is compared to
0.4 L/m3 (3.0 gal/1000 ft3) when these devices are used for panicle removal (Mcllvaine
Company 1974).
Maintenance Problems
Mobile-bed scrubbers are designed to minimize plugging and scale buildup problems through
the constant motion of the packing spheres. However, these problems can still occur at the
scrubber inlet (wet-dry interface) or on the packing suppon grid. Scale buildup in these areas
can cause an uneven airflow distribution through the bed. Uneven airflows result in some
areas of the packing bed having a high gas velocity, while the gas velocity is much lower in
other areas. This can result in a decrease in collection efficiency and in excessive liquid
carryover. Adjusting the inlet sprays can help solve this problem. As with any spray system.
6-7
-------
nozzles can also be a major maintenance problem. Nozzle maintenance is a special concern in
lime or limestone scrubbing systems because of the large quantities of solids present in the
recycled scrubbing liquor.
Deterioration of the spheres can also be a problem. Neither plastic nor marble balls are
able to withstand high temperatures. The marble cracks and breaks while the plastic deforms.
Most systems have safety mechanisms to prevent a total loss of water that would cause high
temperatures. Deterioration of the balls from constant rubbing against each other can also be
a problem. Glass balls can generally withstand abrasive conditions, whereas plastic balls can-
not; therefore, they wear out quickly.
Summary
Mobile-bed scrubbers are used when high collection efficiency of particulate and gaseous
pollutants is required. Typical applications would be for treating flue gases from industrial
boilers, smelting operations, and kraft pulp mills. The main advantage of mobile-bed scrub-
bers is that they are capable of high-efficiency absorption without plugging. The main disad-
vantage is that they do not efficiently remove particles in the submicron range. A major
maintenance problem is the effect of abrasive wear and high temperatures on packing balls,
causing them to deteriorate.
Mobile-bed scrubbers are generally designed in one stage for panicle collection, or in multi-
ple stages for high-efficiency gas absorption. Gas velocities are much higher than those in
packed or plate towers; therefore, mobile-bed scrubbers can be much smaller in size than
either tower. Because of these high gas velocities, incorporating some type of entrainment
separator is mandatory. Table 6-2 lists some general operating characteristics of mobile-bed
scrubbers.
Table 6-2. Operating characteristics of mobile-bed scrubbers.
Pollutant
Pressure drop
(Ap)
Liquid-to-gas
ratio
(L/G)
Liquid-inlet
pressure
(pi)
Removal
efficiency
(%>
and cut
diameter
Applications
Gases
5-15 cm of water
per stage
(2-6 in. of water
per stage)
2.7-8.0 L/m3
(20-60 gal/1000 ft3)
-
99*% of
theoretical
Mining operations
Pulp mills
Utility boilers
Food industry
Panicles
0.4-0.7 L/m3
(3-5 gal/1000 ft3)
2-3-^m cut
diameter
6-8
-------
Review Exercise
1. Mobile-bed, or moving-bed, scrubbers were developed to
provide the effective mass-transfer characteristics of
or rowers without the plugging
problems.
a. spray (or) venturi
b. packed (or) plate
c. cyclonic (or) orifice
d. ejector (or) spray
2. In mobile-bed scrubbers, particles are collected
a. by using inlet sprays.
b. as they impinge on the wetted surface of the spheres.
c. in a froth, or foam, layer above the bed. c i
d. all of the above
1. b. packed (or) plate
3. True or False? In mobile-bed scrubbers, adding stages or
more packing will usually increase particle collection ^
efficiency.
2. d. all of the above
4. Mobile-bed scrubbers provide the gas absorption efficiency
of packed or plate towers; however, they consume
energy for the same unit operation
a. more
b. less J
c. the same
3. False
5. Gas absorption in mobile-bed scrubbers can be enhanced
by
a. increasing the L/G ratio.
b. adding more packing height.
c. adding stages.
d. all of the above
4. a. more
6. When used for gas absorption, mobile-bed scrubbers
operate at I./C* ratios than when used ro rol-
lect particles. ^
a. much higher 1 v
b. much lower
c. the same
5. d. all of the above
6. a. much higher
6-9
-------
7. Scale buildup or plugging at the mobile-bed scrubber inlet
ran cause rhat leads to a decrease in
efficiency.
a. a low liquid pH //
b. uneven gas flow distribution through the bed
c. excessive liquid carryover
d. low liquid flow
8. In mobile-bed scrubbers, the moving packing is made of
a. glass.
b. plastic.
c. marble. ^
d. any of the above
7. b. uneven gas flow dis-
tribution through
the bed
9. The biggest maintenance problem with mobile-bed scrub-
bers is ball deterioration due to
a. abrasive wear.
b. high temperatures.
c. both high temperatures and abrasive wear.
d. none of the above
8. d. any of the above
10. True or False? A major limitation of mobile-bed scrubbers
is that they are not effective in removing submicron-
sized particles. \
9. c. both high tempera-
tures and abrasive
wear.
11. In mobile-bed scrubbers, gas velocities are much
lower/higher than in packed towers or plate towers;
therefore, mobile-bed scrubbers can be much
Smaller/larger in size.
10. True
11. higher,
smaller
6-10
-------
Baffle Spray Scrubbers
Baffle spray scrubbers are very similar to spray towers in design and operation. However, in
addition to using the energy provided by the spray nozzles, baffles are added to allow the gas
stream to atomize some liquid as it passes over them. A simple baffle scrubber system is shown
in Figure 6-5. Liquid sprays capture pollutants and also remove collected particles from the
baffles. Adding baffles slightly increases the pressure drop of the system.
Particle Collection
These devices are used much the same as spray towers —to preclean or remove particles larger
than 10 fim in diameter. However, they will tend to plug or corrode if particle concentration
of the exhaust gas stream is high.
Gas Collection
Even though these devices are not specifically used for gas collection, they are capable of a
small amount of gas absorption because of their large wetted surface.
Mist eliminator
Baffle
Figure 6-5. Baffle spray scrubber
6-11
-------
Summary
These devices are most commonly used as precleaners to remove large particles (>10 jim in
diameter). The pressure drops across baffle scrubbers are usually low, but so are the collection
efficiencies. Maintenance problems are minimal. The main problem is the buildup of solids
on the baffles. Table 6-3 summarizes the operating characteristics of baffle spray scrubbers.
Table 6-3. Operating characteristics of baffle spray scrubbers.
Pollutant
Pressure drop
(Ap)
Liquid-to-gas
ratio
(L/G)
Liquid-inlet
pressure
(pi)
Removal
efficiency
(%)
and cut
diameter
Applications
Gases
2.5-7.5 cm of water
(1-3 in. of water)
0.13 L/m3
(1 gal/1000 ft5)
< 100 kPa
(<15 psig)
Very low
Mining operations
Incineration
Chemical process
industries
Particles
10-/mi cut
diameter
Mechanically Aided Scrubbers
In addition to using liquid sprays or the exhaust stream, energy can be supplied to a scrub-
bing system by using a motor. The motor is used to drive a rotor or paddles which, in turn,
generate water droplets for gas and particle collection. Systems designed in this manner have
an advantage of requiring less space than do other scrubbers, but the overall power require-
ments tend to be higher than for other scrubbers of equivalent efficiency. This point might
appear to contradict the contact power principle; however, significant power losses occur in
driving the rotor. Power is not expended to provide for gas-liquid contact.
Fewer mechanically aided scrubber designs are available than are liquid- and gas-phase
contacting collector designs. Two will be discussed here: centrifugal-fan scrubbers and
mechanically induced spray scrubbers.
A centrifugal-fan scrubber can serve as both an air mover and a collection device. Figure
6-6 shows such a system, where water is sprayed onto the fan blades cocurrently with the mov-
ing exhaust gas. Some gaseous pollutants and particles are initially removed as they pass over
the liquid sprays. The liquid droplets then impact on the blades to create smaller droplets for
additional collection targets. Collection can also take place on the liquid film that forms on
the fan blades. The rotating blades force the liquid (and any particles) off of the blades. The
liquid droplets separate from the gas stream because of their centrifugal motion.
6-12
-------
Spray nozzle
Rotor
Figure 6-6. Centrifugal-fan scrubber.
Centrifugal-fan collectors are the most compact of the wet scrubbers since the fan and col-
lector comprise a combined unit. No internal pressure loss occurs across the scrubber, but a
power loss equivalent to a pressure drop of 10.2 to 15.2 cm (4 to 6 in.) of water occurs
because the blower efficiency is low.
Another mechanically aided scrubber, the induced-spray, consists of a whirling rotor
submerged in a pool of liquid. The whirling rotor produces a fine droplet spray. By moving
the process gas through the spray, particles and gaseous pollutants can subsequently be col-
lected. Figure 6-7 shows an induced-spray scrubber that uses a vertical-spray rotor.
Particle Collection
Mechanically aided scrubbers are capable of high collection efficiencies for particles with
diameters of 1 /im or greater. However, achieving these high efficiencies usually requires a
greater energy input than those of other scrubbers operating at similar efficiencies. In
mechanically aided scrubbers, the majority of particle collection occurs in the liquid droplets
formed by the rotating blades or rotor.
Gas Collection
Mechanically aided scrubbers are generally not used for gas absorption. The contact time
between the gas and liquid phases is very short, limiting absorption. For gas removal, several
other scrubbing systems provide much better removal per unit of energy consumed.
6-13
-------
Mist eliminator
Rotor
Figure 6-7. Mechanically induced spray scrubber.
Maintenance Problems
As with almost any device, the addition of moving pans leads to an increase in potential
maintenance problems. Mechanically aided scrubbers have higher maintenance costs than do
other wet collector systems. The moving parts are particularly susceptible to corrosion and
fouling. In addition, rotating parts are subject to vibration-induced fatigue or wear, causing
them to become unbalanced. Corrosion-resistant materials for these scrubbers are very expen-
sive; therefore, these devices are not used in applications where corrosion or sticky materials
could cause problems.
Summary
Mechanically aided scrubbers have been used to control exhaust streams containing par-
ticulate matter. They have the advantage of being smaller than most other scrubbing systems,
since the fan is incorporated into the scrubber. In addition, they operate with low liquid-to-
gas ratios. The disadvantages are their generally high maintenance requirements, low absorp-
tion efficiency, and high operating costs. The performance characteristics of mechanically
aided scrubbers are given in Table 6-4.
6-14
-------
Table 6-4. Operating characteristics of mechanically aided scrubbers.
Pollutant
Pressure drop
(Ap)
Liquid-to-gas
ratio
(L/G)
Liquid-inlet
pressure
(p /.)
Cut
diameter
Applications
Particles
10-20 cm of water
(4.0-8.0 in. of water)
0.07-0.2 L/m3
(centrifugal)
0.5-1.5 gal/1000 ft3
(centrifugal)
20-60 psig
(centrifugal)
< l-/*m cut
diameter
Mining operations
Food product
industries
Chemical industry
Foundries and
steel mills
0.5-0.7 L/m3
(spray rotor)
4-5 gal/1000 ft3
(spray rotor)
Note: These devices are used mainly for particle collection; however, they can also remove gaseous pollutants
that are present in the exhaust stream.
Review Exercise
1. Adding baffles in a spray tower will generally help increase
the particle removal efficiency, but also increases the
a. L/G ratio.
b. pressure drop. [q
c. height of the unit.
d. all of the above
2. Spray towers and baffle spray towers are generally not
effective in removing particles smaller than
a. 10 fj.m.
b. 50 |im.
c. 100 fim.
d. any of the above
1. b. pressure drop.
3. Mechanically aided scrubbers use a rotor to generate water
droplets. These devices usuallv require less
than other scrubbers, hut have rhar rend rn he
higher.
a. liquid, gas flows
b. space, power requirements k
c. power, liquid requirements
2. a. 10 jim.
4. True or False? Mechanically aided scrubbers can serve as
both an air mover and a collection device.
3. b. space,
power requirements
\
4. True
6-15
-------
5. In mechanically aided scrubbers, the majority of particle
collection occurs
a. in liquid droplets formed by the rotating blades.
b. on the wetted blades.
c. at the inlet sprays.
6. True or False? Mechanically aided scrubbers are generally
not used for gas absorption, since several other designs
provide better removal.
5. a. in liquid droplets
formed by the
rotating blades.
7 Mechanically aided srnihhers frenerallv have
maintenance costs than do other wet collectors because of
their moving parts.
a. lower ^
b. higher 1.
c. the same
6. True
8. True or False? Mechanically aided scrubbers operate at
lower liquid-to-gas ratios than do most other scrubbers.
7. b. higher
i
i
\
8. True
References
Beachler, D. S., andjahnke, J. A. October 1981. Control of particulate emissions. APTI
Course 413, EPA 450/2-81-066. U.S. Environmental Protection Agency.
Bethea, R. M. 1978. Air pollution control technology. New York: Van Nostrand
Reinhold Co.
Environmental Protection Agency (EPA). 1969. Control techniques for particulate air
pollutants. AP-51.
Joseph, G. T., and Beachler, D. S. December 1981. Control of gaseous emissions. APTI
Course 415, EPA 450/2-81-005. U.S. Environmental Protection Agency.
Mcllvaine Company. 1974. The wet scrubber handbook. Northbrook, IL.
Strauss, W. 1975. Industrial gas cleaning. Oxford: Pergamon Press, Inc.
6-16
-------
Lesson 7
Equipment Associated with
Scrubbing Systems
Lesson Goal and Objectives
Goal
To familiarize you with the operation of equipment associated with scrubbing systems.
Objectives
Upon completing this lesson, you should be able to —
1. distinguish between forced- and induced-draft fans,
2. briefly describe the operation of a centrifugal fan,
3. list two maintenance problems associated with fans, pumps, ducts, and pipes in wet
scrubbing systems,
4. list three types of pipes used in scrubbing systems and the advantages and disadvantages
of each,
5. briefly describe quenchers, and
6. list five important variables that should be monitored in wet scrubbing systems.
Introduction
Many components comprise a complete scrubbing system. To fully appreciate the operation of
a scrubber, it is important to have a basic understanding of all the components of the system.
For instance, fans and ducts are required to transport exhaust gas while pumps, nozzles, and
pipes transport liquid to and from the scrubbing vessel. Water-recirculation and mist-
elimination systems are also necessary. Failure of any of these parts will cause problems for the
entire scrubbing system. This lesson presents an overview of the equipment associated with
scrubbing systems — covering their operation and some typical maintenance problems.
Transport Equipment for Exhaust Gases
and Scrubbing Liquids
Fans transport exhaust gas through ducts to and from the scrubber, while pumps transport
liquids through pipes. Although not part of the scrubber chamber, they are essential to its
operation.
7-1
-------
Fans
Fans used in wet scrubbing systems are usually centrifugal. In centrifugal fans, exhaust gas is
introduced into the center of a revolving wheel, or rotor, and exits at a right angle (90°) to
the rotation of the blades (Figure 7-1). Centrifugal fans are classified by the type and shape of
blades used in the fan. The forward-curved fans use blades that are curved toward the direc-
tion of the wheel rotation. The blades are smaller and spaced closer together than are the
blades in other centrifugal fans. These fans are not usually used if the flue gas contains dust
or sticky materials. They have been used for heating, ventilating, and air conditioning
applications in industrial plants. Backward-curved fans use blades that are curved away from
the direction of wheel rotation. The blades will clog when the fan is used to move flue gas
containing dust and sticky fumes. They may be used on the clean-air discharge of air pollu-
tion control devices or to provide clean combustion air for boilers. Radial fans use straight
blades that are attached to the wheel of the rotor. These fans are built for high mechanical
strength and can be easily repaired. Fan blades may be constructed of alloys or coated steel to
help prevent deterioration when handling abrasive and corrosive exhaust gas. Radial fans are
used most frequently for air pollution control applications; however, backward-curved fans are
also used on wet scrubbing systems. Airfoil fans use thick teardrop-shaped blades that are
curved away from the wheel rotation. Airfoil fans can clog when handling dust or sticky
materials.
Backward-curved
Forward curved
Radial
Airfoil
Figure 7-1. Centrifugal fans.
7-2
-------
Fans used for wet scrubbing systems can be located before or after the scrubber. When
located before the scrubber, they are referred to as forced-draft, positive-pressure, or dirty-side
fans. These fans normally move dry air, but can move moist air depending on process condi-
tions. They are subject to abrasion and solids buildup when dust concentration is high. Abra-
sion on the fan can be reduced by using special wear-resistant alloys, by using replaceable
liners on the wheel, or by reducing fan speed (using a large fan that moves slower). The solids
buildup can sometimes be controlled by using a spray wash to periodically clean the wheel. If
dirty-side fans are used, a cyclone or knockout chamber can reduce dust concentration.
Fans located after the scrubber are always operated wet, and are called induced-draft,
negative-pressure, or clean-side fans. These fans are subject to corrosion and solids buildup
from mist escaping from the entrainment separator. Corrosion problems can result when the
exhaust gas contains acid-forming or soluble electrolytic compounds, especially if the
temperature of the gas stream falls below the dew point of these compounds. Corrosion can be
reduced by using proper construction materials and careful pH control in the scrubbing
system. Solids buildup can occur when the mist escaping from the entrainment separator con-
tains dissolved or settleable solids. As the mist enters the fan, evaporation occurs and some
solids deposit on the wheel. If the buildup on the wheel is uniform, no problems occur until
the buildup starts to flake off, knocking the fan out of balance (Wechselblatt 1975). Keeping
entrainment separators operating efficiently or using clean water sprays on the fan blades will
help reduce solids-buildup problems.
Ducts
Ducts, or ductwork, transport exhaust gas to and from the scrubber. Ducts are carefully
designed to keep pressure losses at a minimum. In general, this requires sizing the duct prop-
erly and minimizing the number of bends, expansions, and contractions. Sizing the duct to
suit the exhaust stream velocity will generally reduce the amount of dust that settles in the
ductwork. Bends, expansions, and contractions cause pressure losses in the system and, conse-
quently, increase operating costs.
Abrasion and corrosion are common problems of ductwork. Abrasion is generally more
severe on ductwork leading into the scrubber, while corrosion affects ductwork leaving the
scrubber. Using proper construction materials or linings greatly reduces corrosion or abrasion.
For example, ductwork can be lined partially or fully with brick (especially at elbows) to pre-
vent erosion due to abrasion. For ductwork exiting the scrubber, special alloys resistant to acid
attack should be used. Also, ductwork can be insulated to prevent acids in the flue gas from
condensing.
Pumps
A wide variety of pumps are used to transport both the scrubbing liquid and the sludge. The
proper choice of a pump depends on flow rate, pressure, temperature, and material being
pumped. Electric-motor-driven centrifugal pumps are the pumps most frequently used in wet
scrubbing systems (Calvert et al. 1972). Figure 7-2 illustrates a simple centrifugal pump. The
rotating impeller produces a reduction in pressure at the eye (center) of the impeller, causing
liquid to flow into the impeller from the suction pipe. The liquid is then forced outward along
the blades and discharged.
7-3
-------
Figure 7-2. Centrifugal pump.
As with fans, abrasion and corrosion are the major maintenance problems associated with
pumps in scrubbing systems. The impellers, housing, and seals are subject to potential corro-
sion and abrasion problems. Abrasion is caused by solids buildup in the scrubbing liquid.
Bleeding this liquid and removing the solids before recycling it back through the pump (or
scrubber) will reduce pump wear. Most vendors suggest that the solids content be less than
15% (EPA 1982). Special alloys or rubber linings can also be used to help reduce abrasion
and corrosion.
Pipes
Pipes transport liquid to and from the scrubber. As with pumps, pipes are susceptible to abra-
sion, corrosion, and plugging. A wide variety of materials can be used to make pipes to
reduce these problems. Some advantages and disadvantages of pipe materials commonly used
are given in Table 7-1.
To prevent solids from building up in or plugging the pipe, a liquid slurry velocity in the
scrubbing system of 1.2 to 2.1 m/s (4 to 7 ft/sec) is recommended as a reasonable compromise
(Czuchra 1979).
7-4
-------
Table 7-1. Pipe materials for scrubber systems—advantages and disadvantages.
Material
Advantages
Disadvantages
Metals
Cast iron
Flanged, threaded, or welded
Not resistant to corrosion
Steel
Inexpensive
Stainless steel
Easy to cut and install on site
Copper alloys
Linings used with metal pipes
Hard rubber
Good resistance to many strong
acids and alkalis
Cannot be cut to size on site
Soft rubber
Resists abrasion
Must be precisely manufactured
Glass
Resists acid and alkali attack
Fragile
Thermoplastic
Resists corrosion
Not as abrasion resistant as
PVC
Easily site-installed
rubber or stainless steel
Polyethylene
Good resistance to temperature
Polypropylene
and stress
Nonmetals
Plastic
Resists corrosion
May not be as heat resistant as
other materials
Fiberglass-reinforced pipe (FRP)
Resists chemical corrosion
Less abrasion resistant than
On-site installation
rubber-lined pipe
Operates at higher temperatures
than a solid plastic pipe
Adapted from Calvert et al. 1972.
Conditioning Equipment for Exhaust Gases
Quenchers
Occasionally hot exhaust gas is quenched by water sprays before entering the scrubber. This
can be accomplished by spraying liquid into the exhaust gas. Hot gases (those above ambient
temperature) are often cooled to near the saturation level by sprays before they enter a scrub-
ber. If not cooled, the hot gas stream can evaporate a large portion of the scrubbing liquor,
adversely affecting collection efficiency. Some liquid droplets can evaporate before they have a
chance to contact pollutants in the exhaust stream, and others can evaporate after contact,
causing captured particles to become reentrained. In some cases, quenching can actually save
money. Cooling the gases reduces the temperature and, therefore, the volume of gases, per-
mitting the use of less expensive materials of construction and a smaller scrubber vessel and
fan.
Quenchers are designed using the same principles as scrubbers. Increasing the gas-liquid
contact in them increases their operation efficiency. Small liquid droplets cool the exhaust
stream quicker than large droplets because they evaporate more easily. Therefore, less liquid
is required. However, in most scrubbing systems, approximately one-and-a-half to two-and-a
half times the theoretical evaporation demand is required to ensure proper cooling (Industrial
Gas Cleaning Institute 1975). Evaporation also depends on time —it does not occur instan-
taneously. Therefore, the quencher should be sized to allow for an adequate exhaust-stream
residence time. Normal residence times range from 0.15 to 0.25 seconds for gases under
540°C (1000°F) to 0.2 to 0.3 seconds for gases hotter than 540°C (Schifftner 1979).
7-5
-------
The cleanest water available should be used for presaturating. Quenching with recirculated
scrubber liquor can reduce overall scrubber performance, since recycled liquid usually con-
tains a high level of suspended and dissolved solids. As the liquid droplets evaporate, these
solids become reentrained in the exhaust gas stream. Dissolved solids in the evaporating
quench liquid can form fine particles that are difficult to collect in the scrubber (Kalika
1969). To help reduce this problem, makeup water can be added directly to the quench
system rather than by adding all makeup water to a common sump (EPA 1982).
Construction Materials
By now it should be obvious that scrubbing systems require special materials to prevent or
reduce corrosion and abrasion. These are summarized in Table 7-2.
Table 7-2. Construction materials for wet scrubber components.
Material
Properties/uses
Corrosion resistance
Metal
Cast iron
Carbon steel
Martensitic stainless steel
(410, 416, 420, 440c)
Ferricic stainless steel
405
430
442, 446
High strength; low ductility;
brittleness; hardness; low cost
Good strength, ductility, and
workability; low cost
Chromium alloy, hardenable by
heat treatment; typically used for
machine pans; costs 2 to 5 times
more than carbon steel
Chromium alloy, not hardenable
by heat treatment; costs 2 to 4
times more than carbon steel
Modified for weldability
General purpose, often used for
chimney liners
Used in high-temperature service
Ordinary cast irons exhibit fair
resistance to mildly corrosive
environments; high-silicon cast
irons exhibit excellent resis-
tance in a variety of environ-
ments (hydrofluoric acid is an
important exception); cast irons
are susceptible to galvanic cor-
rosion when coupled to copper
alloys or stainless steels
Fair to poor in many environ-
ments; low pH and/or high dis-
solved solids in moist or immer-
sion service leads to corrosion;
properly applied protective coat-
ings give appropriate protection
in many applications; susceptible
to galvanic corrosion when
coupled to copper alloys or stain-
less steels
Good
Good; better than martensitic
stainless steel; resists stress corro-
sion; better chloride resistance
than austenitic stainless steels
Good resistance to atmospheric
corrosion
7-6
-------
Table 7-2. Construction materials for wet scrubber components (continued).
Material
Properties/ uses
Corrosion resistance
Austenitic stainless steel
Chromium and nickel alloy; not
Excellent; better than martensitic
hardenable by heat; hardenable
or ferritic stainless steel (except
by cold working; nonmagnetic
for halides)
Types 201, 202, 301, 302, 303,
304, and 304L cost 3 to 5 times
more than carbon steel; types
310, 316, 316L, and 321 cost 4 to
10 times more than carbon steel
201, 202
Nitrogen added, used as a substi-
tute for 301 and 302
301
Good hardenability
302
General purpose
304
General purpose
304 L
Modified for weldability
310
Used in high-temperature service
316
Used in corrosive environments
Superior corrosion resistance;
316L
Improved weldability
good acid resistance; resistant to
hot organic acids; good pitting
resistance
Nickel alloy
Good strength; costs over 10 times
Excellent resistance in most
more than carbon steel
environments; not resistant in
strong oxidizing solutions such
as ammonium and HNOj
Inconel® °
Good resistance to stress corrosion
Monel® "
Good resistance to hydrofluoric
acid
Hastelloy® b and
Excellent overall resistance
Chlorimet® c
Titanium
High strength; light weight (60%
Exceptional resistance at ambient
that of steel); costs over 10 times
temperatures; excellent resis-
more than carbon steel
tance at other temperatures,
except that crevice corrosion is
possible in chloride solutions
above 110°C (250°F)
"Registered trademark of Huntington Alloys, Inc.
'Registered trademark of the Stalite Division of Cabot Corporation.
'Registered trademark of the Duriron Company, Inc.
7-7
-------
Table 7-2. Construction materials for wet scrubber components (continued).
Material
Properties/uses
Corrosion resistance
Nonmetal
Glass and glass linings
Brick linings
Carbon brick
Acid brick
Silicon carbide brick
Porcelain and stoneware
Rubber
Plastics
Brittleness, subject to damage by
thermal shock; can be protected
against breakage by coating with
polyester fiberglass
Used when fluorides are present;
540 °C (1000°F) temperature
limit
870 °C (1600°F) temperature limit
1370°C (2500 °F) temperature
limit; high installation costs
Same properties but greater
strength than glass; easily dam-
aged by thermal shock
Excellent mechanical properties
and abrasion resistance; temper-
ture limit of approximately
105 °C (220°F)
Less resistance to mechanical
abuse, lower strength, and
higher expansion rates; cannot
be used where temperatures con-
stantly exceed 105 °C (220 °F)
Good resistance to hydrochloric
and dilute sulfuric acid
Acid resistant and abrasion
resistant; also provides thermal
protection for inner materials
Good acid resistance
Resistant to dilute acids, alkalis,
and salts, but some oxidizing
media will attach to it
Excellent resistance to weak acids
and alkalis; do not corrode and
are not affected by slight changes
in pH or oxygen content
Sources: EPA 1982 and Perry 1973.
Monitoring Equipment
Having adequate equipment is imperative when monitoring the performance of a scrubber.
Instrumentation on a wet scrubber can provide three distinct services:
• obtaining operational information by recording daily data to help detect any problems or
misoperation that may occur,
• providing operating input for other devices to automatically operate some pans of the
system, and
• providing for safety by sounding alarms and/or releasing interlocks to protect both the
operators and equipment.
A monitoring system must be properly installed and maintained to provide reliable data.
Monitors should be installed, operated, and calibrated according to the manufacturer's
instructions. This is essential in obtaining reliable information. Because every scrubbing svstem
is unique, the instrumentation and variables measured will vary from source to source. Table
7-3 lists monitors that are typically used in wet scrubbing systems.
7-8
-------
Table 7-3. Monitoring equipment for wet scrubbing systems.
Monitor
Measurements
Thermometer or
thermocouple
Measures inlet and outlet tempera-
tures of gas to and from scrubber
Measures inlet and outlet tempera-
tures of liquid to and from scrubber
Flowmeter
Measures liquid flow rate to scrubber
Measures the amount of recycled
liquid and bleed stream
Measures flow rate of fresh makeup
liquid to scrubber
Manometer
Measures pressure drop (inlet and
outlet static pressure) across fan,
scrubber vessel, and entrainment
separator
pH meter
Measures pH level in chemical feed
stream, scrubbing liquid, recycle
liquor, and bleed stream
Ammeter
Monitors the current of the fans and
pumps
For any of these monitors, high and/or low settings can be chosen so that if the set value is
exceeded, an alarm sounds, a bypass is opened, or an emergency system is activated. For
example, sources scrubbing hot gases normally have a high-temperature alarm and/or an
interlock system to automatically introduce emergency water or to bypass the scrubber if the
high-temperature setting is exceeded.
7-9
-------
Review Exercise
1. What is/are the most popular type(s) of centrifugal fan for
wet scrubbing systems?
a. radial fans
b. backward-curved fans
c. vane-axial fans s
d. a and b only
e. all of the above
2. Fans located before the scrubber are referred to as
fans.
a. positive-pressure .
b. dirty-side ^-4,
c. forced-draft
d. all of the above
1. d. a and b only
3. Fans located after the scrubber are always operated
a. wet. x
b. dry. ^
2. d. all of the above
4. What is/are the primary maintenance problem(s) associ-
ated with fans?
a. abrasion N
b. solids buildup C.l
c. corrosion
d. all of the above
3. a. wet.
5. True or False? In general, electric-motor-driven centrifugal
pumps are the most frequently used pumps in wet scrub-
bing systems. ^ \
4. d. all of the above
6. What area(s) of the pump is/are most susceptible to abra-
sion or corrosion?
a. impeller
b. housing
c. seals
d. all of the above ^
5. True
7. To reduce pressure losses in ducts, the number of
should he kppr rn a minimum
a. bends
\
b. expansions y-\
c. contractions O7
d. all of the above
6. d. all of the above
7. d. all of the above
7-10
-------
8. What is a/are common problem(s) for pipes in most
scrubbing systems?
a. abrasion
b. corrosion \
c. plugging /v>
d. all of the above
9. True or False? Cast iron and steel pipes are very resistant
to attack by corrosive materials. ^
8. d. all of the above
10. The the liquid droplet produced bv the
quench spray, the more efficient the quencher is in cooling
the exhaust gas stream.
a. smaller
b. larger 'g
c. rounder
d. heavier
9. False
11. Quenchers must be sized to allow for an adequate
of the exhaust gas, sinre evaporation does rint
occur instantaneously. \
10. a. smaller
12. Ouenching should be done with the water X
available.
a. dirtiest
b. cleanest t )
c. highest-pH
d. lowest-pH
11. residence time
13. List five monitors used in scrubbing systems.
12. b. cleanest
i
i
13. • thermometer or
thermocouple
• manometer
• pH meter
• flowmeter
• ammeter
7-11
-------
References
Calvert, S.; Goldshmid, J.; Leith, D.; and Mehta, D. August 1972. Wet scrubber system
study, volume I: scrubber handbook. EPA-R2-72-118a. U.S. Environmental Protection
Agency.
Calvert, S.; Jadmani, I. L.; Young, S.; and Stahlberg, S. October 1974. Entrainment separa-
tors for scrubbers—initial report. EPA 650/2-74-119a. U.S. Environmental Protection
Agency.
Czuchra, P. A. April 1979. Operation and maintenance of a particulate scrubber system's
ancillary components. Presented for the U.S. EPA Environmental Research Information
Seminar, at Atlanta, GA.
Environmental Protection Agency. September 1982. Control techniques for particulate emis-
sions from stationary sources—volume I. EPA 450/3-81-005a.
Gleason, T. G. 1977. How to avoid scrubber corrosion. In Air pollution control and design
handbook. P. N. Cheremisinoff and R. A. Young, eds. New York: Marcel Dekker, Inc.
Industrial Gas Cleaning Institute. 1975. Scrubber system major auxiliaries. Publication WS-4.
Stamford, CT.
Joseph, G. T., and Beachler, D. S. December 1981. Control of gaseous emissions. APTI
Course 415, EPA 450/2-81-005. U.S. Environmental Protection Agency.
Kalika, P. W. 1969. How water recirculation and steam plumes influence scrubber design.
Chem. Eng. 79:133-138.
Kashdan, E. R., and Ranada, M. B. January 1979. Design guidelines for an optimum scrubber
system. EPA 600/7-79-018. U.S. Environmental Protection Agency.
MacDonald, J. W. 1982. Absorbers. In Air pollution control equipment, design, selection,
operation and maintenance. L. Theodore and A. J. Buonicore, eds. Englewood Cliffs:
Prentice-Hall, Inc.
National Asphalt Pavement Association. 1978. The maintenance and operation of exhaust
systems in the hot mix batch plant. Information Series 52, 2nd ed.
Perry, J. H., ed. 1973. Chemical engineers' handbook, 5th ed. New York: McGraw-Hill
Book Co.
Schifftner, K. C. April 1979. Venturi scrubber operation and maintenance. Presented for
the U.S. EPA Environmental Research Information Center, at Atlanta, GA.
Wechselblatt, P. M. 1975. Wet scrubbers (particulates). In Handbook for the operation and
maintenance of air pollution control equipment. F. L. Cross and H. E. Hesketh, eds.
Westport: Technomic Publishing Co., Inc.
7-12
-------
Lesson 8
Wet Flue Gas Desulfurization Systems
Lesson Goal and Objectives
Goal
To familiarize you with the operation of flue gas desulfurization (FGD) systems that use a
scrubbing liquid to absorb S02 present in the exhaust gas stream.
Objectives
Upon completing this lesson, you should be able to —
1. briefly describe five FGD wet scrubbing processes — four nonregenerable and one
regenerable,
2. list six operating variables that affect wet scrubber operation in FGD systems,
3. recognize operating problems associated with each process above, and
4. recall some of the various scrubber designs and operating conditions associated with FGD
processes.
Introduction
One of the largest markets for wet scrubbing systems, in terms of money spent, is flue gas
desulfurization (FGD). FGD refers to the removal of S02 from the process exhaust stream.
The majority of FGD systems have been applied to combustion sources such as utility and
some industrial coal-fired boilers. FGD systems are also used to reduce S02 emissions from
some industrial plants such as smelters, acid plants, refineries, and pulp and paper mills.
FGD systems can be operated wet or dry. (Since dry systems do not incorporate wet scrub-
bers, they will not be discussed in this lesson.) In wet scrubbing systems, liquid absorbs S02 in
the exhaust stream. The scrubbing liquid contains an alkali reagent to enhance S02 absorp-
tion. More than a dozen different reagents have been used, with lime and limestone being the
most popular for utility boilers, and sodium-based reagents the most popular for industrial
boilers (Table 8-1). Sodium-based solutions (sometimes referred to as clear solutions) provide
better S02 solubility and less scaling problems than lime or limestone. However, sodium
reagents are much more expensive. Wet FGD scrubbers can further be classified as
nonregenerable or regenerable. Nonregenerable processes, sometimes called throwawav
processes, produce a sludge waste that must be disposed of properly. Most regenerable
processes produce a product that mav be sold to partially offset the cost of operating the FGD
svstem. Regenerated products include elemental sulfur, and sulfuric acid. Based on the recent
capacities listed in Table 3-1, 95% of che FGD processes are nonregenerable. or throwawav.
The throwawav processes are simpler and presentlv more economical to use than those used to
recover and sell products.
8-1
-------
Table 8-1. FGD processes.
Operational
Construction
Planned
rrocess
No.
MW
No.
MW
No.
MW
Utility boilers—throwaway
Dual alkali
3
1181
0
0
2
842
Dual alkali/limestone
1
20
0
0
0
0
Lime
22
8801
4
1995
11
6841
Limestone
32
11.464
17
7637
29
17,088
Limestone/alkaline fly ash
2
1480
0
0
0
0
Lime/alkaline fly ash
9
2613
2
1400
0
0
Lime/limestone
2
20
0
0
0
0
Lime/ spray drying
1
110
3
1060
5
1813
Sodium carbonate
4
925
1
330
4
1900
Sodium carbonate/spray drying
0
0
1
440
0
0
Utility boilers—regenerable
Aqueous carbonate/spray drying
0
0
1
100
0
0
Citrate
1
60
0
0
0
0
Lime
0
0
1
65
0
0
Limestone
0
0
0
0
0
0
Lime/limestone
0
0
0
0
0
0
Magnesium oxide
0
0
3
724
0 '
0
Wellman-Lord
7
1540
1
534
0
0
Industrial boilers—throwawav
Caustic wastestream
5
520
0
0
1
78
Double alkali
7
361
2
193
3
204
Lime/limestone
2
48
1
2
0
0
Sodium carbonate
9
1284
2
110
2
93
Sodium hydroxide
13
397
0
0
3
180
Sources: Smith et al. 1981 and Tuttle et al. 1979.
Note: The data for utility boilers are for 1980.
The data for industrial boilers are for 1978.
Most FGD systems employ two stages: one for fly ash removal and the other for S02
removal. Attempts have been made to remove both the fly ash and S02 in one scrubbing
vessel; however, these systems experienced severe maintenance problems and low simultaneous
removal efficiencies. The flue gas normally passes first through a fly ash removal device, either
an electrostatic precipitator or a wet scrubber, and then into the S02 absorber.
Many different types of absorbers have been used in FGD systems, including spray towers,
Venturis, plate towers, and mobile packed beds. Because of scale buildup, plugging, or
erosion, which affect FGD dependability and absorber efficiency, the trend is to use simple
scrubbers such as spray towers instead of more complicated ones. The configuration of the
tower may be vertical or horizontal, and flue gas can flow cocurrentlv, countercurrentlv, or
crosscurrently to the liquid. The chief drawback of spray towers is that thev have a higher
liquid-to-gas ratio requirement (for equivalent S02 removal) than other absorber desims
(Makansi 1982).
8-2
-------
Numerous operating variables affect the S02 removal rate, of the absorber. Most of these
variables were discussed in previous lessons; however, some are unique to FGD absorbers. The
following list contains some of the important parameters affecting the operation of an FGD
scrubber (Ponder et al. 1979 and Leivo 1978):
Liquid-to-gas ratio—The ratio of scrubber slurry to gas flow (L/G ratio). For a given set of
system variables, a minimum L/G ratio is required to achieve the desired S02 absorption,
based on the solubility of S02 in the liquid. High L/G ratios require more piping and struc-
tural design considerations, resulting in higher costs.
pH —Depending on the particular type of FGD system, pH must be kept within a certain
range to ensure high solubility of S02 and to prevent scale buildup.
Gas velocity —To minimize equipment cost, scrubbers are designed to operate at maximum
practicable gas velocities, thereby minimizing vessel size. Maximum velocities are dictated
by gas-liquid distribution characteristics and by the maximum allowable liquid entrainment
that the mist eliminator can handle. Gas velocities may be 1.5 to 10 m/s (5 to 30 ft/sec) in
tower scrubbers and more than 30 m/s (100 ft/sec) in the throat of a venturi scrubber. A
common range of the gas velocity for FGD absorbers is 2.0 to 3.0 m/s (7 to 10 ft/sec). The
lower the velocity, the less the entrainment, but the more costly the scrubber will be.
Slurry holdup—For FGD processes using an alkali slurry for scrubbing, the system should
be designed to provide adequate residence time for the S02 to be absorbed by the alkali
slurry. The main objective is to make sure that the maximum amount of alkali is utilized in
the scrubber. Residence times in packed towers may be as long as 5 seconds. Residence
times in venturi scrubbers are a few hundredths of a second, usually too short for high
absorption efficiency of SOs in systems using lime or limestone scrubbing slurries, unless
additives or two scrubbing stages are used.
Gas distribution — A major problem that has occurred in commercial FGD scrubbers is
maintaining a uniform gas flow. If the flow is not uniform, the scrubber will not operate at
design efficiencies. In practice, uniform flow has been difficult to achieve. Typically, turn-
ing vanes near the scrubber inlet duct and compartmentalization have been used.
Scrubber designs —To promote maximum gas-liquid surface area and contact time, a
number of scrubber designs have been used. Common ones are mobile-bed scrubbers,
venturi-rod scrubbers, plate towers, and packed towers. Countercurrent packed towers are
infrequently used because they have a tendency to become plugged by collected particles or
to scale (when lime or limestone scrubbing slurries are used).
Turndown —To adjust to changes in boiler load. The scrubber must provide good gas-
liquid distribution, high liquid holdup for some processes, and high gas-liquid interfacial
area for varying gas flow rates. Some scrubbers can be turned down to 50% of design,
while others must be divided into sections that can be closed off. A variable-throat venturi
can be used to accommodate turndown. In a large FGD installation, individual modules
can be taken out of service.
It is important to note that the above list does not imply that these are the only parameters
affecting S02 absorption efficiency. Each FGD process has a unique set of operating criteria.
8-3
-------
In addition to the set of factors just given, the coal properties greatly affect FGD system
design for boiler operations. The major coal properties affecting FGD system design and
operation are (Leivo 1978):
Heating value of coal —Affects flue gas flow rate. Flow rate is generally higher for lower
heating value coals, which also contribute a greater water-vapor content to the flue gas.
Moisture content — Affects the heating value and contributes directly to the moisture con-
tent and volume of the flue gas.
Sulfur content—The sulfur content, together with the allowable emission standards, deter-
mines the required S02 removal efficiency, the FGD system complexity and cost, and also
affects sulfite oxidation.
Ash content—May affect FGD system chemistry and increase erosion. In some cases, it may
be desirable to remove fly ash upstream from the FGD system.
Chlorine content—May require high-alloy metals or linings for some process equipment
and could affect process chemistry or require prescrubbing.
Another important design consideration associated with most FGD systems is that the flue
gas exiting the absorber is saturated with water and still contains some S02 (no system is
100% efficient). Therefore, these gases are highly corrosive to any downstream equip-
ment—i.e., fans, ducts, and stacks. Two methods used to minimize corrosion are reheating
the gases to above their dew point or choosing construction materials and design conditions
that allow equipment to withstand the corrosive conditions. The selection of a reheating
method or the decision not to reheat (thereby using special construction materials) are very
controversial topics connected with FGD design (Makansi 1982). Both alternatives are expen-
sive and must be considered on a by-site basis.
Four methods are currently used to reheat stack gases:
1. Indirect in-line reheating—The flue gas passes through a heat exchanger that uses
steam or hot water.
2. Indirect-direct reheating—Steam is used to heat air (outside the duct) and then the hot
air is mixed with the scrubbed gases.
3. Direct combustion reheating—Oil or gas is burned either in the duct or in an external
chamber, and the resulting hot gases are mixed with the scrubbed gases.
4. Bypass reheating—A portion of the untreated hot flue gas bypasses the scrubber and is
mixed with the scrubbed gases.
None of the above methods has a clear advantage over the others (Makansi 1982). Systems
using indirect in-line reheating have experienced severe corrosion and plugging problems;
indirect-direct and direct combustion reheating are expensive because of added fuel costs; and
bypass reheating is limited in the degree of reheating obtainable (due to S02 emissions in che
bypass).
This lesson will discuss five of the more popular FGD systems—four throwaway processes
and one regenerable process. The process chemistry, system description, and operating
experience involved in each will be presented.
8-4
-------
Review Exercise
1. -based solurions absorb SO-, better than
: however, the former are mnrh more
expensive.
a. Sodium, lime or limestone i^")
b. Lime or limestone, sodium
c. Gypsum, lime or limestone
d. Limestone, lime
-
2. True or False? Almost all FGD systems use a single wet
scrubber for both SOz and fly ash removal. y"
1. a. Sodium,
lime or limestone
3. Spray rowers require higher
(for equivalent S02 removal) than other absorber
designs.
a. pressure drops ^
b. gas velocities
c. liquid-to-gas ratios
d. all of the above
2. False
Most use two scrubbing
stages.
4. The lower the gas velocity, the more/less;the entrap-
ment; however, the scrubber system will^be more/less
i ''—
costly.
3. c. liquid-to-gas ratios
5. List five properties of the coal (or fuel) that will affect
FGD operation.
4. less,
more
6. Because flue gas contains some SOj as it exits the ;
absorbers. FGD svstems generally use to pre-
vent corrosion.
a. additional absorbers
b. reheaters
c. special construction materials for downstream fans and
ductwork
d. both b and c
5. • heating value
• sulfur content
• chlorine content
• ash content
• moisture content
7. Match the reheating method with the proper description.
1. Indirect in-line ^a. A portion of the hot
2. Indirect-direct - \ \ untreated flue gas is mixed
3. Direct combustion - j ) I with the scrubbed gases.
4. Bypass -— // b. Flue gas passes through a
/ / heat exchanger in the duct.
/ '-¦-c. Oil or gas is burned, and the
! hot gases are mixed with the
V scrubbed gases.
d. Steam is used to heat air, and
this hot air is mixed with the
flue gas.
6. d. both b and c
8-5
-------
7. 1. b
2. d
3. c
4. a
Nonregenerable FGD Processes
Nonregenerable FGD processes generate a sludge or waste product. The sludge must be dis-
posed of properly in a pond or landfill. The three most common nonregenerable processes
used on utility boilers in the U.S. are lime, limestone, and double-alkali. Although the
double-alkali process regenerates the scrubbing reagent, it is classified as throwaway since it
does not produce a salable product and generates solids that must be disposed of in a landfill.
Sodium-based throwaway systems (NaOH and Na2C03) are overwhelmingly chosen for
industrial boilers (see Table 8-1).
Lime Scrubbing
Process Chemistry
Lime scrubbing uses an alkaline slurry made by adding lime (CaO), usually 90% pure, to
water. The alkaline slurry is sprayed in the absorber and reacts with the S02 in the flue gas.
Insoluble calcium sulfite (CaSOs) and calcium sulfate (CaSO*) salts are formed in the
chemical reaction that occurs in the scrubber and are removed as sludge.
A number of reactions take place in the absorber. Before the calcium can react with the
S02, both must be broken down into their respective ions. This is accomplished by slaking
(dissolving) the lime in water and then spraying the slurry into the flue gas to dissolve the S02.
Simplified reactions occur simultaneously and are illustrated below.
SOt dissociation: Lime (CaO) dissolution:
SOjjioaou,) ~*S02(a?UWM) "H H20 * Ca( OH)2 (aqueous)
S02 + H20-H2S03 Ca(0H2)~Ca~ + 20H-
H2S03—H*+ HSO; —2H*+ SO3
Now that the S02 and lime are broken into their ions (SOj and Ca~), the following reaction
occurs:
Ca~ + SO, + 2H" + 20H~ —CaS03 + 2HzO.
In addition, the following reactions can also occur when there is excess oxygen:
SOj+ i/602-S0;
SO4 + Ca4"*—CaS04(JO/1<<).
From the above relationships and assuming that the lime is 90% pure, it will take 1.1 moles
of lime to remove 1 mole of S02 gas.
8-6
-------
System Description
The equipment necessary for S02 emission reduction comes under four operations:
1. Scrubbing or absorption —accomplished with scrubbers, holding tanks, liquid-spray
nozzles, and circulation pumps.
2. Lime handling and slurry preparation — accomplished with lime unloading and storage
equipment, lime processing and slurry preparation equipment.
3. Sludge processing—accomplished with sludge clarifiers for dewatering, sludge pumps and
handling equipment, and sludge solidifying equipment.
4. Flue-gas handling—accomplished with inlet and outlet ductwork, dampers, fans, and
stack gas reheaters.
Figure 8-1 is a schematic of a typical lime FGD system. Flue gas from the boiler first passes
through a particulate emission removal device then into the absorber where the S02 is
removed. The gas then passes through the entrainment separator to a reheater and is finally
exhausted out of the stack. Individual FGD systems vary considerably, depending on the FGD
vendor and the plant layout. ESPs or scrubbers can be used for particle removal with the
various absorbers used for S02 removal.
A slurry of spent scrubbing liquid and sludge from the absorber then goes to a recirculation
tank. From this tank, a fixed amount of the slurry is bled off to process the sludge, and, at
the same time, an equal amount of fresh lime is added to the recirculation tank. Sludge is
sent to a clarifier, where a large portion of water is removed from the sludge and sent to a
holding tank. Makeup water is added to the process-water holding tank, and this liquid is
returned to the recirculation tank. The partially dewatered sludge from the clarifier is sent to
a vacuum filter, where most of the water is removed (and sent to the process-water holding
tank) and the sludge is sent to a settling pond. Table 8-2 lists operational data of lime FGD
systems, showing the various absorbers used.
Reheater
Absorber with 'iliiilr
entrainment
separator Makeup
water
Particle
scrubber
Vacuum filter
Clarifier
Recirculation
tank
Process-water
holding tank
Settling pond or landfill
Figure 8-1. Typical process flow for a lime or limestone FGD system.
8-7
-------
Table
8-2. Operational data
for lime FGD systems on utility boilers.
Company and plant
MW
FGD vendor
Fly ash control
%S
SO, absorber
Number
of modules
L/G ratio
Pressure drop
(Ap)
Efficiency 1
(%)
(gross)
name
coal
per boiler
L/m'
gal/1000 ft*
kPa
in. HtO
Design
Test
Pennsylvania Power
Bruce Mansfield H1
Bruce Mansfield #2
Bruce Mansfield #3
917
917
917
Chemico
Chemico
Pullman Kellogg
lst-stage venturi
lst-stage venturi
ESP
3.0
3.0
3.0
Fixed-throat venturi
Fixed-throat venturi
Weir crosscurrent spray
6
6
6
6.0
6.0
45.0
45.0
2.0
2.0
0.7
8.0
8.0
2.8
92.1
92.1
92.0
95.0
95.0
95.0
Columbus & Southern
Ohio Electric
Concsvillc ttb
Conesvillc #6
411
411
Air Correction Division
Air Correction Division
ESP
ESP
4.67
4.67
Mobile bed
Mobile bed
1
2
6.7
6.7
50.0
50.0
2.0
2.0
8.0
8.0
89.5
89.5
89.7
89.5
Duquesne Light
Eh ama
I'lullips
510
408
Chemico
Chemico
ESP
Cyclone/ESP
2.20
1.92
Variable-throat venturi
(plumb-bob type)
Variable-throat venturi
5
4
5.3
5.3
40.0
40.0
4.0
4.0
16.0
16.0
83.0
83.0
86.0
90.0
Kentucky Utilities
Green River
64
American Air Filter
Cyclone/
variable-throat
venturi
4.0
Mobile bed
1
4.5
34.0
1.0
4.0
80.0
80.0
I^ouisville Gas &
Electric
Cane Run tt'l
Cane Run ttb
Mill Creek #1
Mill Cicek #3
Paddy's Run #6
188
200
358
442
72
American Air Filter
Combustion Engineering
Combustion Engineering
American Air Kilter
Combustion Engineering
ESP
ESP
ESP
ESP
3.75
3.75
3.75
3.75
2.50
Mobile bed
Countercurrent spray
Mobile bed
Mobile bed (marbles)
2
2
4
2
8.0
7.4
12.7
8.7
2.2
60.0
55.0
95.0
65.0
16.5
1.0
0.1
1.6
2.9
40.0
0.5
6.5
11.5
85.0
85.0
85.0
85.0
90.0
87.5
91.0
86.6
85.7
90.0
Kansas City Power
& Light
Hawthorn tt'i
I la wi hoi n #4
90
90
Combustion Engineering
Combustion Engineering
-
0.6
0.6
Mobile bed (marbles)
Mobile bed (marbles)
2
2
3.5
3.5
26.0
26.0
2.7
2.7
11.0
11.0
70.0
70.0
70.0
70,0
Monongahela Power
Pleasants #1
Pleasants #2
618
618
B&W
B&W
ESP
ESP
3.7
4.5
Sieve tray
Sieve tray
4
4
7.4
7.4
55.0
55.0
1.2
5.0
90.0
90.0
90.0
90.0
Utah Power & Light
1 iuiiter It 1
Hunter #2
Iiumingion tt 1
400
400
430
Chemico
Chemico
Cheniico
ESP
ESP
ESP
0.55
0.55
0.55
Countercurrent spray
Countercurrent spray
Countercurrent spray
4
4
4
5.7
5.7
5.7
43.0
43.0
43.0
0.6
0.6
0.6
2.5
2.5
2.5
80.0
80.0
80.0
80.0
80.0
80.0
Note: A dash ( ) ind
icuics tl
*«i[ no daia air available:
-------
Operating Experience
Early lime FGD systems were plagued with many operational and maintenance problems.
Scale buildup and plugging of absorber internals and associated equipment were prominent
problems. However, scaling and plugging in lime FGD systems were not as severe as with
other calcium-based FGD systems (EPA 1981). Scale buildup (CaS04) on spray nozzles and
entrainment separators was particularly troublesome. New spray nozzle designs and careful
control of the recirculating slurry have reduced internal scrubber scaling (EPA 1975). Prob-
lems with the entrainment separators have also been reduced by careful separator design,
installing adequate wash sprays, and monitoring the pressure drop across them. Additional
techniques used to reduce scale buildup are (Leivo 1978):
Control of pH — If a lime FGD system is operated above a pH of 8.0 to 9.0, there is a risk
of sulfite scaling. Automatic contol of the feed by on-line pH sensors has been successful.
Holding tank residence time—By providing retention time in the scrubber recirculation
tank, the supersaturation of the liquor can be decreased before recycling to the scrubber.
Typical residence times of 5 to 15 minutes have been used in some full-scale systems.
Control of suspended solids concentration — The degree of supersaturation can be
minimized by keeping an adequate supply of seed crystals in the scrubber slurry. Typical
levels in newer installations range from 5 to 15% suspended solids. Solids are generally con-
trolled by regulating the slurry bleed rate.
Liquid-to-gas ratio—High liquid-to-gas ratios can reduce scaling problems because the
absorber outlet slurry is more dilute, containing less calcium sulfates and calcium sulfites
that cause scaling.
Another problem concerned stack gas reheaters. Stack gas is reheated to avoid condensa-
tion on and corrosion of the ductwork and stack, and to enhance plume rise and pollutant
dispersion. Reheating is accomplished by using steam coils in the stack, by using hot air
supplied by auxiliary oil heaters in the stack, or by other methods previously mentioned.
Some reheater failures were caused by acid attack to reheater components. Other reheaters
vibrated too much, causing structural deterioration.
Corrosion of scrubber internals, fans and ductwork, and stack linings have been reduced by
using special materials such as rubber- or plastic-coated steel and by carefully controlling
slurry pH with monitors. Additional operation and maintenance problems and solutions are
found in Proceedings: Symposium on Flue Gas Desulfurization, Volumes I and II (EPA:
March 1978, July 1979, and 1981), and Lime FGD Systems Data Book (Ponder et al. 1979).
8-9
-------
Review Exercise
1. List three nonregenerable FGD processes.
2. Reacting lime with water is referred to as
a. clarifying.
b. slaking.
c. raking.
d. thickening.
1. • lime
• limestone
• double-alkali
What is CaS03 in the following reaction?
Ca~ + SO3 + 2H* + 20H~—CaS03 + 2H20
a. solid (sludge)
b. liquid
c. gas
2. b. slaking.
4. Lime FGD systems use a(an)
to remove fly ash
3. a. solid (sludge)
from the flue gas before it enters the absorber.
a. venturi scrubber
b. electrostatic precipitator
c. mechanical collector with precipitator or scrubber
d. any of the above
5. Most lime FGD systems on utility boilers operate at L/G
ratios of
a. 0.4 to 1.3 L/m3 (3 to 10 gal/1000 ft3).
b. 3.0 to 8.0 L/m3 (25 to 60 gal/1000 ft3).
c. 13 to 26 L/m3 (100 to 200 gal/1000 ft3).
d. none of the above
4. d. any of the above
6. In early lime FGD systems, scale buildup and plugging of
the were particularly troublesome.
a. spray nozzles
b. entrainment separator
c. scrubber internals
d. all of the above
5. b. 3.0 to 8.0 L/m3
(25 to 60 gal/1000 ft3).
o
7. Operating a lime FGD system at a pH above 8.0 to 9.0
a. reduces scale buildup.
b. increases the risk of scale buildup.
c. is recommended. '(-J
d. eliminates nozzle plugging.
6. d. all of the above
High/Low liquid-to-gas ratios reduce the potential for
scale buildup.
7. b. increases the risk of
scale buildup.
8. High
8-10
-------
9. Stack gas is reheated to
a. avoid condensation.
b. enhance plume rise. i
c. give better pollutant dispersion. (1^
d. all of the above
9. d. all of the above
Limestone Scrubbing
Process Chemistry
Limestone scrubbers are very similar to lime scrubbers. The use of limestone (CaC03) instead
of lime requires different feed preparation equipment and higher liquid-to-gas ratios (since
limestone is less reactive than lime). Even with these differences, the processes are so similar
that an FGD system can be set up to use either lime or limestone in the scrubbing liquid.
The basic chemical reactions occurring in the limestone process are very similar to those in
the lime-scrubbing process. The only difference is in the dissolution reaction that generates the
calcium ion. When limestone is mixed with water, the following reaction occurs:
CaCO, {solid) + H20 -Ca~ + HCO; + OH".
The other reactions are the same as those for lime scrubbing.
System Description
The equipment necessary for S02 absorption is the same as that for lime scrubbing, except in
the slurry preparation. The limestone feed (rock) is reduced in size by crushing it in a ball
mill. Limestone is sent to a size classifier. Pieces larger than 200 mesh are sent back to the
ball mill for recrushing. Limestone is mixed with water in a slurry supply tank. Limestone is
generally a little cheaper than lime, making it more popular for use in large FGD systems.
Table 8-3 lists operations data for limestone FGD systems. Note the similarities in equipment
and operating conditions to those of lime FGD systems.
8-11
-------
Table 8-3. Operational data for limestone FCD systems on utility boilers.
Company and plant
MW
FGI) vendor
Fly ash control
%S
in
SO, absorber
Number
of modules
L/G ratio
Pressure drop
(Ap)
Efficiency
(%)
(gross)
name
coal
per boiler
L/m"
gal/1000 ft*
kPa
in. H,0
Design
Test
Alabama Electric
Tombigbee #2
Tombigbee #3
255
255
Peabody
Peabody
ESP
ESP
1.15
1.15
Countercurrent spray
Countercurrent spray
2
2
9.4
9.4
70.0
70.0
1.0
1.0
4.0
4.0
59.5
59.5
85.0
85.0
Arizona Electric Power
Apache #2
Apache #3
Choila #1
Cholla #2
195
195
119
264
Research Cottrell
Research Cot 11 ell
Research Cottrell
Research Cottrell
ESP
ESP
Cyclone/venturi
Cyclone/venturi
0.50
0.50
0.50
0.50
Spray/packed bed
Spray/packed bed
Spray/packed bed
Spray/packed bed
2
2
1
4
2.8
2.8
6.5
6.5
20.6
20.6
48.9
48.9
1.5
1.5
0.1
0.1
6.0
6.0
0.5
0.5
42.5
42.5
58.5
75.0
97.0
97.0
92.0
85.0
Basin Electric Power
Laramie River #1
Laramie River #2
570
570
Research Cottrell
Research Cottrell
ESP
ESP
0.81
0.81
Spray/packed bed
Spray/packed bed
5
5
8.0
8.0
60.0
60.0
-
-
90.0
90.0
90.0
90.0
(Central Illinois Light
Duck Creek #1
416
Emironeering
ESP
3.66
Rod deck packed tower
4
6.7
50.0
2.0
8.0
85.0
85.0
Colorado Ute
Electrical
Craig #1
Craig #2
447
455
Peahody
Peabody
ESP
ESP
0.45
0.45
Countercurrent spray
Countercurrent spray
4
4
6.7
6.7
50.0
50.0
1.6
1.6
6.5
6.5
85.0
85.0
85.0
85.0
Common wealth Edison
Powei ion
450
Air Correction Division —
UOP
ESP
S.53
Mobile bed (TCA)
3
8.0
60.0
3.0
12.0
74.0
75.5
Indianapolis Power &
Light
Petersburg #3
532
Air Correction Division —
UOP
ESP
3.25
Mobile bed (TCA)
4
6.7
50.0
1.7
7.0
85.0
85.0
Kansas City Power 8c
Light
I,a Cygne
Jctiieiy # 1
Jclfery #2
Lawrence #4
Lawrence Hb
820
720
700
125
420
B&W
Combustion Engineering
Combustion Engineering
Combustion Engineering
Combustion Engineering
Variable venturi
ESP
ESP
Rod venturi
Rod venturi
5.39
0.32
0.30
0.55
0.55
Sieve tray
Countercurrent spray
Countercurrent spray
Countercurrent spray
Countercurrent spray
8
6
2
2
5.0
4.1
4.1
4.0
2.5
37.7
30.4
30.4
30.0
19.0
1.5
1.0
1.0
0.6
0.6
6.0
6.0
6.0
2.5
2.5
80.0
80.0
80.0
73.0
52.0
80.0
60 0
60.0
73.0
52 0
Note: A dash ( - ) indicates lliat 110 data are available.
-------
Table 8-3. Operational data for limestone FGD systems on utility boilers (continued).
Company and plant
MW
FG1) vendor
Fly ash control
%S
in
SO, absorber
Number
of modules
L/G ratio
Pressure drop
(Ap)
Efficiency
<%)
(gross)
name
coal
per boiler
L/m*
gal/1000 ft'
kPa
in. HtO
Design
Test
Salt River Project
Coronado #1
Coronado #2
350
350
Pullman Kellogg
Pullman Kellogg
ESP
ESP
LOO
1.00
Weir crosscurrent spray
Weir crosscurrent spray
2
2
-
-
0.4
0.4
1.5
1.5
66.0
66.0
82 0
82.0
South Carolina Public
Service
Winyah #2
Winyah #3
280
280
B&W
B&W
ESP
ESP
1.70
1.70
Venturi/sieve tray
Countercurrent spray
2
2
6.3
47.5
1.1
4.5
45.0
90.0
90.0
90.0
South Mississippi
Electric
R. D. Morrow //I
K. D. Morrow Ut
200
200
Environed ing
Enviioneci ing
ESP
ESP
1.30
1.30
Rod deck packed tower
Rod deck packed tower
1
1
6.6
6.6
49.0
49.0
2.0
2.0
8.0
8.0
52.7
52.7
85.0
85.0
Southern Illinois
Marion #4
173
B&W
ESP
3.75
Countercurrent spray
2
9.9
74.0
1.5
6.0
89.4
89.4
Springfield City
Southwest #1
194
Air Correction Division —
UOP
ESP
3.50
Mobile bed (TCA)
2
5.5
41.0
1.5
6.0
80.0
87.0
Springfield Water,
Light & Power
Dallman U'i
205
Reseaich-Cottrell
Cyclone/ESP
3.30
Spray/packed bed
2
0.2
0.7
95.0
95.0
TV A
Widows Creek #8
550
TV A
ESP/venturi
3.70
Mobile packed bed
and grid packing
1
3
8.0
60.0
0.5
2.0
70.0
Texas Power & Light
Sandow #4
545
Combustion Engineering
ESP
1.60
Countercurrent spray
3
—
—
—
—
75.0
—
Texas Utilities
Mania Lake HI
Martin Lake #2
Mania Lake //3
Moniicello
793
793
793
800
Research- Cottrell
Research Courell
Research Coltrell
Chemico
ESP
ESP
ESP
ESP
0.90
0.90
0.90
1.50
Spray/packed bed
Spray/packed bed
Spray/packed bed
Countercurrent spray
6
6
6
3
9.4
70.0
1.1
1.1
1.1
1.2
4.5
4.5
4.5
5.0
71.0
71.0
71.0
74.0
95.0
95.0
95.0
74.0
-------
Operating Experience
Early limestone FGD systems had scrubber operating problems similar to those of lime scrub-
bing systems. Plugged and clogged nozzles, scrubber internals, and mist eliminators (entrain-
ment separators) resulted from inefficient S02 absorption by limestone in the scrubber.
Increased absorption efficiency is achievable at high pH values since more alkali is available to
dissolve the S02 gas. However, scale buildup will occur if the scrubber is operated at very high
pH values. The pH levels can be maintained by carefully controlling limestone and water feed
rates. Low pH reduces removal efficiency; high pH causes scale buildup on scrubber internals.
As can be seen from Tables 8-2 and 8-3, the S02 removal efficiencies for various lime and
limestone FGD installations range from 50 to 92%. These FGD systems were designed to meet
existing air pollution regulations. Lime and limestone FGD systems are capable of removing
S02 with efficiencies in excess of 90% (Devitt et al., March 1978; EPA 600/7-78-032a). The
addition of small amounts of soluble magnesium (< 1 % by weight) to the scrubber liquor can
greatly increase S02 removal efficiencies to as high as 99% (Dewitt et al., March 1978; EPA
600/7-78-032a). Magnesium is added in the form of magnesium oxide, magnesium sulfate, or
dolomitic lime (used in lime scrubbing systems). Magnesium compounds are more soluble than
calcium compounds and react rapidly with S02.
EPA is currently working on a program that uses an additive of adipic acid to limestone
FGD systems. Adipic acid can increase S02 removal efficiencies from 85% to as high as 97%
(EPA, August 1980). Adipic acid is a crystalline powder derived from petroleum. EPA
experiments have shown that when a limestone slurry reacts with S02 in the scrubber, the
slurry becomes very acidic. This acidity limits the amount of SO* that can be absorbed.
Adding adipic acid to the slurry slightly increases the slurry's initial acidity, but prevents it
from becoming highly acidic during the absorption of S02. The net result is an improvement
in the scrubbing efficiency.
EPA research has shown that adipic acid can reduce the total limestone consumption by as
much as 15%, thus reducing operating costs. Adipic acid is nontoxic (it is used as a food
additive) and does not have any adverse environmental impacts. Adipic acid has been tested
in full-scale tests at an electrical generating station, and its benefits have been verified
(Mobley and Chang 1981).
Another scrubber operating problem occurring in lime and limestone FGD systems is that
calcium sulfite in the sludge settles and filters poorly. It can be removed from the scrubber
slurry only in a semiliquid or paste-like form. A process improvement called forced oxidation
was developed by IERL-RTP to address this problem. In forced oxidation, air is blown into
the scrubber slurry tank that contains primarily calcium sulfite and water. The air oxidizes
the calcium sulfite to calcium sulfate.
CaSO, + H,0 4- 14 02 -CaSO< + H20
Calcium sulfate formed by this reaction grows to a larger crystal size than calcium sulfite. As
a result, calcium sulfate is easily filtered, forming a drier and more stable material that can
be disposed of in a landfill. This material (CaS04) can also be used for cement, gypsum
wallboard, or as a fertilizer additive.
Forced oxidation can also help control scale buildup problems on scrubber internals. This
process helps control scale by removing calcium sulfite from the slurry in the form of calcium
sulfate, which is more easily filtered. This will prevent calcium sulfites and calcium sulfates
from being recirculated in the absorber. However, if forced oxidation is used on a closed-loop
water system, there is a potential for increasing the concentrations of chlorides and other
impurities in the recycled water that previously were thrown away with the sludge.
8-14
-------
Review Exercise
1. Limestone FGD systems generally operate at higher'/lower
liquid-to-gas ratios than lime FGD systems because SOz is
more/less reactive with a limestone slurry.
2. True or False? The chemistry for S02 removal in a lime-
stone slurry is very different from that for S02 remfoyal in
a lime slurry. (
1. higher,
less
3. The major difference in equipment for a limestone FGD
system (compared to a lime FGD system) is in the
a. fly ash collection equipment.
b. type of absorber.
c. slurry feed preparation.
d. all of the above
2. False
4. True or False? Limestone is generally less expensive to pur-
chase than is lime. '"T
3. c. slurry feed
preparation.
5 The addition of to the slurry of a limestone
FGD system has increased S02 removal rates in experimen-
tal studies.
a. magnesium oxide j
b. magnesium sulfate
c. adipic acid
d. any of the above
4. True
6. In lime/limestone FGD systems, calcium sulfite formed as
pan of the sludge is difficult to remove from the slurry.
One method used to eliminate this problem is to convert
the calcium sulfite to calcium sulfate by the process called
a. forced oxidation.
b. Wellman-Lord. ' ;
c. double-alkali.
d. direct reduction.
5. d. any of the above
6. a. forced oxidation.
Dical-Alkali Scrubbing
Dual-, or double-, alkali scrubbing is a throwaway FGD process that uses a sodium based
alkali solution to remove S02 from combustion exhaust gas. The sodium alkali solution absorbs
S02, and the spent absorbing liquor is regenerated with lime or limestone. Calcium sulfites
and sulfates are precipitated and discarded as sludge. The regenerated sodium scrubbing solu-
tion is returned to the absorber loop. The dual-alkali process has reduced plugging and
scaling problems in the absorber because sodium scrubbing compounds are very soluble. Dual-
alkali systems are capable of 95% S02 reduction.
8-15
-------
Particulate matter is removed prior to S02 scrubbing by an electrostatic precipitator or a
venturi scrubber. This is done to prevent fly ash erosion of the absorber internals and to pre-
vent any appreciable oxidation of the sodium solution in the absorber due to catalytic
elements in the fly ash (EPA, March 1978).
Process Chemistry
The sodium alkali solution is usually a mixture of sodium carbonate (Na2C03), also called
soda ash, sodium sulfite (Na2S03), and sodium hydroxide (NaOH), also called caustic. The
S02 reacts with the alkaline components to primarily form sodium sulfite and sodium bisulfite
(NaHS03). The following are the main absorption reactions (EPA 1981):
2 NaOH + S02 - Na2S03 + H20
NaOH + S02 —NaHS03
Na2C03 + S02 + H20 ~ 2NaHS03
Na2C03 + S02 — Na2S03 + C02
Na2S03 + S02 + H20-2NaHS03.
In addition to the above reactions, some of the S03 present may react with alkaline com-
ponents to produce sodium sulfate. For example,
2NaOH + S03 - Na2SO« + H20
Throughout the system, some sodium sulfite is oxidized to sulfate by:
2Na2S03 + 02 —2Na2S04.
After reaction in the absorber, spent scrubbing liquor is bled to a reactor tank for regenera-
tion. Sodium bisulfite and sodium sulfate are inactive salts and do not absorb any S02.
Actually, it is the hydroxide ion (OH"), sulfite ion (S03), and carbonate ion (C03) that absorb
S02 gas. Sodium bisulfite and sodium sulfate are reacted with lime or limestone to produce a
calcium sludge and a regenerated sodium solution.
2NaHS03 + Ca(OH)2 - Na2S03 + CaS03 • H201 + 3/2H20
(lime) (sludge)
Na2S03 + Ca(OH)2 + i/2 H20 - 2NaOH + CaS03 • 4 H20
(lime) (sludge)
Na2S04 + Ca(OH)2 - 2NaOH + CaS041
(lime) (sludge)
At the present time, lime regeneration is the only process that has been used on commercial
dual-alkali installations.
System Description
The dual-alkali process uses two loops —absorption and regeneration. In the absorption loop,
the sodium solution contacts the flue gas in the absorber to remove S02. As shown in
Figure 8-2, the scrubbing liquor from the bottom of the absorber is mixed with regenerated
solution and sprayed in at the top of the absorber. A bleed stream from the recirculating
liquid is sent to the reactor tank in the regeneration loop. The bleed stream is mixed with a
8-16
-------
Stack
Clarifier Makeup water
Absorber
Mixing tank
Vacuum filter
To landfill
Figure 8-2. Typical process flow for a dual-alkali FGD system.
lime slurry in a reactor tank, where insoluble calcium salts are formed and the absorbent is
regenerated. The sludge from the reactor is then sent to a clarifier, or thickener, where the
calcium sludge is drawn off the bottom, filtered, and washed with water. From the filter, the
sodium solution is recycled to the thickener, and the sludge is discarded. From the thickener,
the regenerated sodium solution is sent to a holding tank where the sodium compounds and
makeup water are added.
Some sodium sulfate solution is unreacted in the regeneration step. Additional sodium is
added to the regenerated solution in the form of soda ash or caustic soda. This regenerated
absorbent is now ready to be used again.
Operating Experience
The dual-alkali process has been installed and operating on both utility and industrial boilers
for a number of years. It is the third most popular FGD system used on industrial boilers (see
Table 8-1). Corrosion of, erosion of, and scale buildup on system equipment have not been
major operating problems at dual-alkali FGD installations in the U.S. (EPA 1981). Operating
data for the dual-alkali systems are presented in Table 8-4. Note the much lower L/G ratios
of these systems compared to those of lime and limestone systems. This is because the sodium
solution is more efficient in absorbing SO, than are either lime or limestone slurries.
8-17
-------
Tabic 8-4. Operational data for dual-alkali FCD systems on utility and industrial boilers.
Company and plant
MW
FGI) vendor
Fly ash control
%S
in
SO, absorber
Number
of modules
L/G ratio
Pressure drop
(Ap)
Efficiency
(%)
(gross)
name
coal
per boiler
L/m*
gal/1000 ft*
kPa
in. H.O
Design
Test
(Central Illinois
Public Service
Newton tfl
617
Biu'll L'.nvnoit-'t'li
ESP
2.50
Mobile bed
4
1.3
10.0
1.5
6.0
90.0
90.0
I.ouisville Gas &
Electric
Cant Run #6
299
Combustion Equipment
Asm>i:iju;s
ESP
4.80
Sieve plates
2
1.3
10.0
2.5
9.9
95.0
'.)•> .0
Southern Indiana
Gas 8c Electric
A. B. Biown ttl
265
FMC
ESP
3.55
Disc and donut trays
2
1.3
10.0
2.5
10.0
85.0
85.0
Caterpillar Tractor
East i'eoria, 11.
Joliet, IL
Morton, 11.
Mossvillc, IL
105
34
19
70
KMC
Zurn
Zurn
FMC
Cyclone
Cyclone
Cyclone
Cyclone
3.20
3.20
3.20
Venturi
Dustraxtor
Dustraxtor
Venturi
4
2
2
4
2.2
1.2
16.0
8.6
-
-
-
90.0
90.0
90.0
90.0'
Firestone Tire
Pottstown. l'A
4
FMC
Cyclone
3.00
Venturi
1
1.3
10.0
_
—
90.5
General Motors
I'ainia, OH
32
CM Environmental
Cyclone
—
Bubble-cap plates
4
2.6
20.0
0.9
8.0
—
90.0
Note: A dash ( ) indicates that no data are available.
-------
Some operating problems include regenerating scrubbing liquor and controlling the solids
content of the sludge. Sodium sulfate, one of the compounds in the spent scrubbing liquor, is
difficult to regenerate because it does not react efficiently with hydrated lime in the presence
of sodium sulfite (Leivo 1978). Process conditions must be carefully controlled to adjust for
the amounts of sodium sulfate and sodium sulfite that are formed in the spent scrubbing
liquid. Another problem occurring in dual-alkali systems is that the solids content of the
sludge can vary greatly, causing problems in handling and stabilizing the sludge for final
disposal (Makansi 1982).
Review Exercise
1 . Dual-alkali processes frenerallv use a snlurinn
to absorb the S02 from the flue gas and then react it with
a slurrv to regenerate the absorbing solution.
a. sodium, citrate
b. citrate, lime or limestone
c. sodium, lime or limestone
d. lime or limestone, sodium
2. SO, is /mor^/less soluble in a sodium alkali solution than
v—y .—
in a lime or limestone slurry.
1. c. sodium,
lime or limestone
3. True or False? The sodium solution used in FGD systems is
often a mixture of different sodium compounds. T
2. more
4. Solutions of sodium compounds are referred to as clear
liquor solutions because the compounds are
a. blue. *
b. soluble. 'Vj
c. insoluble.
d. transparent.
3. True
5. In the dual-alkali process, the sodium reagent is regen-
erated by reacting the spent solution with lime. As part of
this reaction, insoluble are formed in the
regeneration vessel.
a. sodium salts
b. calcium salts \
c. magnesium salts
d. citrate salts
4. b. soluble.
5. b. calcium salts
8-19
-------
6. Compared to lime and limestone scrubbing systems,
dual-alkali absorbers have a much lower
a. pressure drop. ^
b. gas velocity. (
c. liquid-to-gas ratio.
d. all of the above
7. True or False? Using sodium-based scrubbing solutions
(as compared to calcium-based) helps eliminate scale \
buildup.
6. c. liquid-to-gas ratio.
7. True
Sodium-Based Once-Through Scrubbing
Sodium-based once-through (throwaway) scrubbing systems are the overwhelming choice for
FGD systems installed on industrial boilers (see Table 8-1). These systems use a clear liquid
absorbent of either sodium carbonate, sodium hydroxide, or sodium bicarbonate. According
to Makansi (1982), sodium-based systems are favored for treating flue gas from industrial
boilers because:
• sodium alkali is the most efficient of the commercial reagents in removing S02, and the
chemistry is relatively simple.
• they are soluble systems —as opposed to slurry systems —making for scale-free operation
and fewer components.
• such systems can handle the wider variations in flue-gas composition resulting from the
burning of many different fuels by industry.
• the systems are often smaller, and operating costs are a small percentage of total plant
costs.
• in some cases, these plants have a waste caustic stream or soda ash available for use as
the absorbent.
These systems have been applied to only a few large utility boilers because:
• the process consumes a premium chemical (NaOH or Na2C03) that is much more costly
per pound than calcium-based reagents.
• the liquid wastes contain highly soluble sodium salt compounds. Therefore, the huge
quantities of liquid wastes generated by large utilities would have to be sent to ponds to
allow the water to evaporate.
Process Chemistry
The process chemistry is very similar to that of the dual-alkali process, except the absorbent is
not regenerated.
8-20
-------
System Description
A basic sodium-based throwaway FGD system is illustrated in Figure 8-3. Exhaust gas from
the boiler may first pass through an ESP or baghouse to remove particulate matter. Sodium
chemicals are mixed with water and sprayed into the absorber. The solution reacts with the
S02 in the flue gas to form sodium sulfite, sodium bisulfite, and a very small amount of
sodium sulfate. A bleed stream is taken from the scrubbing liquor recirculation stream at a
rate equal to the amount of S02 that is being absorbed. The bleed stream is sent to a
neutralization tank and aeration tower before being sent to a lined disposal pond.
Stack
Fan
Water and
soda ash
^ Absorber
Recirculation
stream
Aeration
tower
Mixing tank
Neutralization tank
Disposal pond
Figure 8-3. Typical process flow for a sodium-based throwaway (single-alkali) FGD system.
Some coal-fired units use ESPs or baghouses to remove fly ash before the gas enters the
scrubber. In these cases, the absorber can be a plate tower or spray tower that provides good
scrubbing efficiency at low pressure drops. For simultaneous S02 and flv ash removal, venturi
scrubbers can be used. In fact, many of the industrial sodium-based throwaway systems are
venturi scrubbers originally designed to remove particulate matter. These units were slightly
modified to inject a sodium-based scrubbing liquor. Although removal of both particles and
S02 in one vessel can be economically attractive, the problems of high pressure drops and
using a scrubbing medium to remove fly ash must be considered. However, in cases where the
particle concentration is low, such as from oil-fired units, simultaneous particulate and SO;
emisson reduction can be effective.
8-21
-------
Operating Experience
Presently, 93 sodium-based throwaway FGD systems are in operation in the U.S. They have
been installed on 158 industrial boilers and 4 utility boilers. Table 8-5 lists operating data for
some of these systems. These systems are generally simpler to operate and maintain than lime
or limestone systems. Therefore, reported operating problems have not been as severe or as
frequent with the sodium-based system as with calcium-based systems.
Control of pH, as with other FGD systems, is of prime concern to maximize absorption effi-
ciency. Troubles with controlling pH can cause scale buildup and plugging of the sample lines
(at high pH, the liquor absorbs C02 and forms carbonate scale in systems where a high
amount of calcium or magnesium is present) (Makansi 1982). Other problems include ineffec-
tive entrainment separation, nozzle plugging, and failure of dampers, duct liners, and stack
liners.
8-22
-------
Table 8-5. Operational data for sodium-based once-through FGD systems on utility and industrial boilers.
Company and plain
name
MW
FGD vendor
Fly ash control
%S
in
SO, absorber
Number
of modules
L/G ratio
Pressure drop
-------
Table 8
1-5. Operational data for sodium-based once-through FGD systems on utility and industrial boilers (continued).
Company and plant
name
MW
(gross)
FGD vendor
Fly ash control
%S
in
coal
SO, absorber
Number
of modules
per boiler
L
/G ratio
Pressure drop
(Ap)
Efficiency
<%)
L/m'
gal/1000 ft'
kPa
in. H,0
Design
Test
Getty Oil
Bakcrslicld, CA
Bakerstield, CA
Oi-cuu, CA
36
445
2:5
KMC
In-house
In house
None
None
None
1.10
1.10
4.00
Disc-and-donut tray/
flexitray
Flexitray
Packed tower
1
9
1
1.1
1.2
8.4
9.0
-
-
...
90.0
96.0
94.0
ITT Raynier
Kernandina Beach, FI.
88
Neptune Airpol
Cyclone
2.50
Variable-throat venturi
2
—
—
5.5
22.0
—
85.0
Kcrr-McGte
Trona, CA
245
Combustion Equipment
Association
-
0.5-5
Plate tower
2
-
-
1.5
6.0
98.0
Mead Paperboard
Stevenson, AL
50
Neptune Airpol
Venturi
3.00
Bubble-cap plates
1
_
—
—
95.0
Northern Ohio Sugar
Fiecnujin, OH
20
Great Western Sugar
None
1.00
Variable-throat venturi
2
—
_
Keichhold Chemicals
lVnsacola, KI-
Texasgulf
Granger, WV
40
Neptune Airpol
None
2.00
Venturi
2
—
—
h II
24.0
—
--
70
Swcinco
Cyclone/ESP
0.75
Sieve plate
2
—
-
-
-
90.0
-------
Review Exercise
1. What is the most popular FGD system for industrial
boilers?
a. lime
b. limestone j.
c. Wellman-Lord C/
d. sodium-based throwaway
2. True or False? The three sodium compounds used most
often in throwaway systems are sodium hydroxide, /
sodium carbonate, and sodium bicarbonate. /
1. d. sodium-based
throwaway
3. Sodium-based once-through FGD systems are favored for
industrial boilers because
a. sodium is the most efficient of the commercial reagents.
b. they operate without scale buildup occurring.
c. they are often smaller and cheaper than other systems.
d. all of the above ; t
k '
2. True
4. Large utilities have not used sodium-based once-through
systems because of the expense of the sodium reagent and
the
a. limited efficiency.
b. low fly ash removal.
c. wastes contain soluble salts that cannot be discharged
into rivers or lakes. r
d. all of the above
3. d. all of the above
5. True or False? In a sodium-based once-through FGD sys-
tem, the flue gas may first pass through a baghouse or
ESP.
i
i
4. c. wastes contain soluble
salts that cannot be
discharged into rivers
or lakes.
6. A problem that must be considered when trying to remove
both S02 and fly ash in the same scrubber is that
a. pressure drops are higher.
b. the scrubbing liquid, if recirculated, can contain a high
level of fly ash. i
c. SOo absorption efficiency is always lower.
d. a. and b. above
5. True
! 6. d. a. and b. above
8-25
-------
7. True or False? Sodium-based once-through systems are
generally simpler to operate and maintain than lime
limestone FGD systems.
6. d. a. and b. above
8. At high pH values, the scrubbing liquid in the sodium
svsrem absorbs and can form carbonate scale
if significant amounts of calcium and magnesium are
present.
a. S02
b. C02 )
c. 02
d. CaC03
7. True
00
cr
n
O
Regenerable FGD Processes
Regenerable FGD processes remove S02 from the flue gas and generate a salable product.
Regenerable products include elemental sulfur, sulfuric acid, or, in the case of lime or
limestone scrubbing, gypsum (used for wallboard). Regenerable processes do not produce a
sludge, thereby eliminating the sludge disposal problem. Most regenerable processes also:
• have the potential for consistently obtaining a high S02 removal efficiency, usually
exceeding 90%,
• utilize the scrubbing reagent more efficiently than nonregenerable processes, and
• use scrubbing liquors that do not cause scaling and plugging problems in the scrubber.
The major drawback of using these processes is that these systems are usually more com-
plicated in design and are more expensive to install and operate.
Two regenerable processes presently operating in the U.S. are the Wellman-Lord and the
magnesium oxide. The Wellman-Lord process has been widely used in both sulfuric acid and
petroleum refining industries and has been installed on a number of industrial and utility
boilers. The magnesium oxide process has been tested at a number of utility boilers, but the
Philadelphia Electric Company's Eddystone and Cromby Stations are the only utility boilers
presently operating this process. The Wellman-Lord process is the only major commercial
regenerable FGD process used in the U.S. and will be the only one covered in detail in this
section. The citrate and magnesium oxide processes are covered in more detail in APTI
Course 415, Control of Gaseous Emissions—Student Manual (EPA 450/2-81-006).
Wellman-Lord,
The Wellman-Lord process is a regenerable FGD process used to reduce S02 emissions from
utility and industrial boilers and produces a usable product. This process is sometimes referred
to as the Wellman-Lord/Allied Chemical process, Allied Chemical referring to the regenera-
tion step.
8-26
-------
Process Chemistry
In the Wellman-Lord process, the S02 is absorbed by an aqueous sodium sulfite solution
which forms a sodium bisulfite solution according to the following equation:
S02 + Na2S03 + H20-2Na2HS03.
Some oxidation occurs in the absorber to form sodium sulfate, which is unreactive with S02
gas.
Na2S03 + V£02-Na2S04
The formation of sodium sulfate depletes the supply of sodium sulfite available for scrubbing.
This can be made up by adding sodium carbonate to the scrubbing slurry to combine with
sodium bisulfite according to the following chemical reaction:
Na2C03 + 2NaHS03 — 2Na2S03 + C021 + H20.
The absorbent is then regenerated by evaporating the water from the bisulfite solution.
2NaHS03 — Na2S03 + HzO + S02(conc#„rrot,d)
The concentrated S02 produced in the regeneration step is then sent to the Allied process for
conversion to elemental sulfur or sulfuric acid.
A typical process flow of one Wellman-Lord system is shown in Figure 8-4. The process
equipment includes an electrostatic precipitator for removing particulate matter; a venturi
scrubber for cooling flue gas and removing S03 and chlorides; an SOz absorber; an
evaporator-crystallizer for regenerating the absorbent; and the Allied Chemical process for
reducing concentrated S02 gas into elemental sulfur or sulfuric acid. The absorber is a plate
tower. S02 gas is scrubbed with a sodium sulfite solution at each plate. A mist eliminator
removes entrained liquid droplets from gas exiting the absorber. There is a direct-fired
natural gas reheating system in the absorber stack to reheat cleaned gas for good dispersion of
the steam plume.
Concentrated S02
to sulfur plant
Stack
Evaporator-crystallizer
Electrostatic
precipitator
Fan
Plate
Makeup water
tower
Venturi
scrubber
Dissolving tank
Absorber
surge tank
To sodium sulfate processing
Purge tank
Figure 8-4. Typical process flow for a Wellman-Lord FGD svstem.
8-27
-------
The solution (sodium bisulfite), collected at the bottom of the absorber, overflows into an
absorber surge tank. This solution is pumped through a filter to remove any collected
particulate matter. A small side-stream is sent to a purge treatment system where sodium
sulfate is removed. The solution is then pumped to the evaporator for regeneration of the
sodium sulfite solution.
The evaporator is a forced-circulation vacuum evaporator. Solution is recirculated in the
evaporator, where low-pressure steam evaporates water from the sodium bisulfite solution.
When sufficient water is removed, sodium sulfite crystals form and precipitate. Concentrated
S02 gas (95% by volume) is removed by the steam. The sodium sulfite crystals form a slurry
that is withdrawn continuously and sent to a dissolving tank, where condensate from the
evaporator is used to dissolve the sodium sulfite crystals into a solution. This solution is
pumped back into the top stage of the absorber (EPA 1977). The water vapor is removed
from the evaporator's overhead S02/H20 vapors by water-cooled condensers. The SOz is com-
pressed by a liquid-ring compressor and sent to the Allied Chemical SOz reduction plant.
Operating Experience
The Wellman-Lord process has ben installed on two 350-MW coal-fired boilers at the Public
Service of New Mexico power plant at San Juan. This system was supplied by Davy Powergas
and is similar to the system supplied on the Mitchell power plant in Indiana. The San Juan
plant uses a five-stage tray tower instead of the three-stage tray tower of the Mitchell plant.
The Wellman-Lord process has also been used to control S02 emissions at Claus tail-gas
plants, sulfuric acid plants, and on industrial boilers. A listing of selected installations is given
in Table 8-6. SOz removal efficiency ranges from 85 to 90%, with a high of 98% on units
installed in Japan (EPA, March 1978; EPA 600/7-78-032b).
8-28
-------
Table 8-6. Wellman-Lord installations in the United States.
Gas volume treated
Initial
Company and location
Feed gas origin
startup
date
mVs
scfm
Olin Corporation
Sulfuric acid plant
20.1
43,000
1970
Paulsboro, NJ
Standard Oil of California
Claus plant
13.4
28,000
1972
El Segundo, CA
Allied Chemical Corporation
Sulfuric acid plant
13.4
28,000
1973
Calumet, IL
Olin Corporation
Sulfuric acid plant
34.8
74,000
1973
Curtis Bay, MD
Standard Oil of California
Claus plant
13.4
28,000
1975
Richmond, CA
Standard Oil of California
Claus plant
13.4
28,000
1975
El Segundo, CA
Standard Oil of California
Claus plant
13.4
28,000
1976
Richmond, CA
Northern Indiana Public
115-MW coal-fired boiler system
105.0
223,000
1977
Service Company
with 80% load factor and
Gary, IN
recovery capacity
Public Service Company of
575-MW coal-fired boiler system,
New Mexico
San Juan Station No. 1
Waterflow, NM
840.0
1,780,000
1978
Public Service Company of
375-MW coal-fired boiler system,
New Mexico
San Juan Station No. 2
Waterflow, NM
Getty Oil Company
60-MW mixed-fuel boiler system,
Delaware City, DE
Delaware City No. 1
Getty Oil Company
60-MW mixed-fuel boiler system.
. 235.0
520,000
1980
Delaware City, DE
Delaware City No. 2
Getty Oil Company
60-MW mixed-fuel boiler system,
Delaware City, DE
Delaware City No. 3
Public Service Company of
550-MW coal-fired boiler system.
New Mexico
San Juan Station No. 3
Waterflow, NM
1121.0
2,400,000
1981
Public Service Company of
550-MW coal-fired boiler system,
New Mexico
San Juan Station No. 4
Waterflow, NM
8-29
-------
Review Exercise
1. Regenerable FGD processes generate a salable product
such as
a. sulfur.
b. sulfuric acid.
c. gypsum. ^
d. all of the above
2. List three advantages that the regenerable process
has over the nonregenerable FGD process.
1. d. all of the above
3. In the Wellman-Lord process, an aqueous solution of
ahsnrhs SO..
a. Na2S03
b. Na2HS03 Vi
c. Na2S04 v
d. CaO
2. • avoids sludge disposal
problem
• utilizes reagent better
• uses clear liquid solu-
tions (reduces scaling)
4. In the regenerating reaction of the Wellman-Lord process,
water is evaporated from the sodium bisulfite solution,
regenerating the sodium sulfite and producing
a. elemental sulfur.
b. concentrated S02 gas. I
c. gypsum. :f \
d. all of the above lv^
3. a. Na2S03
(sodium sulfite)
5. True or False? In the Wellman-Lord process, the flue gas
may pass through a particle removal device before entering
the absorber. ( \
4. b. concentrated S02 gas.
6. In the Wellman-Lord process, a prescrubber is used to
a. humidify the gas stream.
b. remove chlorides. (" ¦
c. both a and b
d. none of the above
5. False
The flue gas must be
pretreated.
6. c. both a and b
8-30
-------
Summary
FGD systems have been installed and operated on many industrial and utility boilers and on
some industrial processes for a number of years. These systems are capable of removing
approximately 70 to 90% of the S02 in the flue gas, depending on the operating conditions of
the system. Some systems have achieved an S02-removal efficiency of greater than 95%. The
most popular FGD systems used on utility boilers are lime and limestone scrubbing. Approx-
imately 75% of the FGD systems installed on utility boilers are either lime or limestone scrub-
bing. The use of dual-alkali systems on utility boilers is attractive because of their ability to
remove S02 very efficiently and because of the reduced scaling problems associated with these
systems. Wellman-Lord FGD systems have been used to reduce S02 emissions from utility and
industrial boilers and from a number of industrial processes. These systems have the advan-
tage of regenerating the scrubbing liquor and producing a salable product instead of a sludge
that can be a disposal problem. However, these systems are more expensive to install and
operate than are lime, limestone, or dual-alkali systems. The throwaway-sodium FGD systems
have been used mostly on industrial boilers. These systems use a sodium scrubbing liquor that
is very efficient in absorbing S02 emissions, but produce liquid wastes that can cause waste
disposal problems. FGD systems used on utility boilers generate large quantities of liquid
wastes. Therefore, throwaway-sodium systems have mainly been used on industrial boilers.
Over the past 15 years, a wealth of material has been written and documented concerning
FGD control technology. The authors of this manual suggest that the readers turn to the
many publications from EPA-IERL concerning this subject, particularly the proceedings from
the FGD symposia sponsored by the EPA.
References
Devitt, T.; Gerstle, R.; Gibbs, L.; Hartman, S.; and Klier, R. March 1978. Flue gas desul-
furization system capabilities for coal fired steam generators, volumes I and II.
EPA 600/7-78-032a and b. U.S. Environmental Protection Agency.
Environmental Protection Agency. April 1981. Control techniques for sulfur oxide emissions
from stationary sources. EPA 450/3-81-004.
Environmental Protection Agency. 1981. Proceedings: symposium on flue gas desulfurization—
Houston, Texas, October 1980, volumes I and II. EPA 600/9-81-019a and b.
Environmental Protection Agency. August 1980. Research summary, controlling sulfur oxides.
EPA 600/8-80-029.
Environmental Protection Agency. July 1979. Proceedings: symposium on flue gas desulfuriza-
tion— Las Vegas, Nevada, March, 1979, volumes I and II. EPA 600/7-79-167a and b.
Environmental Protection Agency. May 1979. Sulfur emission: control technology and waste
management. Decision Series. EPA 600/9-79-019.
Environmental Protection Agency. February 1979. Summary report—sulfur oxides control
technology series: flue gas desulfurization, Wellman-Lord process. EPA 625/8-79-001.
Environmental Protection Agency. March 1978. Proceedings: symposium on flue gas desulfuri-
zation—Hollywood, Florida, November 1977, volumes I and II. EPA 600/7-78-058a and b.
8-31
-------
Environmental Protection Agency. 1977. Wellman-Lord S02 recovery process—flue gas
desulfurization plant. EPA Technology Transfer Capsule Report, First Progress Report.
EPA 625-2-77-011.
Environmental Protection Agency. 1976. Lime/limestone wet-scrubbing test results at the EPA
alkali scrubbing test facility. EPA Technology Transfer Capsule Report, Third Progress
Report. EPA 625/2-76-010.
Environmental Protection Agency. 1975. Lime/limestone wet-scrubbing test results at the EPA
alkali scrubbing test facility. EPA Technology Transfer Capsule Report, Second Progress
Report. EPA 625/2-75-008.
Joseph, G. T., and Beachler, D. S. December 1981. Control of gaseous emissions. APTI
Course 415, EPA 450/2-81-005. U.S. Environmental Protection Agency.
Leivo, C. C. March 1978. Flue gas desulfurization systems: design and operating considera-
tions. Volume II. Technical Report. EPA 600/7-78-030b. U.S. Environmental Protection
Agency.
Makansi, J. October 1982. S02 control: optimizing today's processes for utility and industrial
powerplants. Power.
Mobley, J. D., and Chang, J. C. S. 1981. The adipic acid enhanced limestone flue gas
desulfurization process—an assessment./, of Air Poll. Control Assoc. 31:1249-1253.
Ponder, T. C.; Hartman, J. S.; Drake, H. M.; Klier, R. P.; Master, J. S.; Patkar, A. N.;
Tems, R. D.; and Tuttle, J. April 1979. Lime FGD systems data book. EPA 600/8-79-009.
U.S. Environmental Protection Agency.
Smith, M.; Melia, M.; Gregory, N.; and Scalf, K. January 1981. EPA utility FGD survey:
October-December 1980. Volumes I and II. EPA 600/7-81-012a and b. U.S. Environmental
Protection Agency.
Tuttle, J.; Patkar, A.; Kothari, S.; Osterhout, D.; Heffling, M.; and Eckstein, M. April 1979.
EPA industrial boiler FGD survey: first quarter 1979. EPA 600/7-79-067b. U.S. Environ-
mental Protection Agency.
8-32
-------
Lesson 9
Design Review of Scrubbers Used
for Particulate Pollutants
Lesson Goal and Objectives
Goal
To familiarize you with the factors to be considered when reviewing particulate-pollutant
scrubber design plans for the permit process.
Objectives
Upon completing this lesson, you should be able to —
1. recall at least four important scrubber design factors,
2. estimate the collection efficiency of a venturi scrubber using appropriate equations and
graphs, and
3. use the cut power method to estimate the cut diameter necessary for achieving a
specified collection efficiency.
Introduction
The design of a scrubber involves many factors including space restriction, pollutant collection
efficiency, pressure drop (gas-side), particle size, exhaust gas flow rate, liquid-to-gas ratio, and
many construction details such as using corrosion-resistant materials, baffles, nozzles, venturi
throats, water sprays, packing, plates, orifices, entrainment separators, inlets, and outlets.
These have been discussed in detail in the previous lessons. Officers who review scrubber
design plans for air pollution control agencies should consider these factors during the review
process.
In Lesson 1, and throughout this course, scrubbers are categorized by the manner in which
exhaust gas and liquid are brought into contact. Scrubbers can also be grouped by the kinds
of pollutants they collect: those that are mainly used to collect particulate emissions and those
that are mainly used to collect gaseous emissions.
Because all scrubbers can be used to collect both particulate and gaseous pollutants, the
choice of the most appropriate type can sometimes be difficult. Therefore, this lesson will
point out those scrubber design features that are important when choosing a scrubber to
remove particulate pollutants. This lesson will also look at a few equations that can be used to
estimate pressure drop and collection efficiency.
9-1
-------
/
Wet Scrubbers Used to Remove Particles
Venturi scrubbers are the most popular scrubbers used to remove particulate matter. Other
scrubbers used include cyclonic, orifice, mechanically aided, and spray towers. Typical gas-
side pressure drops and L/G ratios for these devices are given in Table 9-1.
Table 9-1. Ranges of pressure drops and liquid-to-gas (L/G) ratios
for various wet scrubbers.
Scrubber
Pressure drop, Ap
Liquid-to-gas ratio*
kPa
in. H»0
L/ms
gal/1000 ft'
Venturi
1.5-25.0
5.0-100.0
0.4-5.0
3.0-40.0
Spray tower
0.12-0.75
0.5-3.0
0.7-2.7
5.0-20.0
Cyclonic spray
0.4-4.0
1.5-10.0
0.3-1.3
2.0-10.0
Moving bed
(good for removing particulate
and gaseous pollutants)
0.5-6.0
2.0-24.0
0.4-8.0
3.0-60.0
Orifice (self-induced spray)
0.5-4.0
2.0-10.0
0.07-0.7
0.5-5.0
Mechanically aided (fan)
1.0-2.0
4.0-8.0
0.07-0.5
0.5-4.0
•Higher L/G reflects those used for gas absorption.
Wet scrubbers remove particles from an exhaust stream by contacting the particles with
liquid, usually water. A number of factors affect particle removal efficiency, including:
• particle-size distribution
• liquid flow rate
• exhaust gas flow rate
• method of contacting
• pressure drop across the scrubber.
As with gaseous pollutant removal (absorption), efficient particle removal requires contact
between the exhaust stream (containing particles) and the scrubbing liquid. However, panicle
removal occurs instantaneously upon contact with the liquid, whereas efficient absorption
requires a long contact time. Therefore, efficient panicle removal requires high relative
velocities (gas versus liquid velocity).
Estimating Collection Efficiency and Pressure Drop
A number of theories have been developed from basic panicle-movement principles to explain
the action of wet scrubbing systems. Many of these stan from firm scientific concepts, but
yield only qualitative results when predicting collection efficiencies or pressure drops. The
interaction of paniculate matter having a given particle-size distribution with water droplets
having another size distribution is not easy to express in quantitative terms. As a result of this
complexity, experimentally determined parameters are usually needed to approach reality
(Beachler andjahnke 1981).
9-2
-------
4
Wet Scrubbers Used to Remove Particles
Venturi scrubbers are the most popular scrubbers used to remove paniculate matter. Other
scrubbers used include cyclonic, orifice, mechanically aided, and spray towers. Typical gas-
side pressure drops and L/G ratios for these devices are given in Table 9-1.
Table 9-1. Ranges of pressure drops and liquid-to-gas (L/G) ratios
for various wet scrubbers.
Scrubber
Pressure drop, Ap
Liquid-to-gas ratio41
kPa
in. HtO
L/mJ
gal/1000 ft*
Venturi
1.5-25.0
5.0-100.0
0.4-5.0
3.0-40.0
Spray tower
0.12-0.75
0.5-3.0
0.7-2.7
5.0-20.0
Cyclonic spray
0.4-4.0
1.5-10.0
0.3-1.3
2.0-10.0
Moving bed
(good for removing particulate
and gaseous pollutants)
0.5-6.0
2.0-24.0
0.4-8.0
3.0-60.0
Orifice (self-induced spray)
0.5-4.0
2.0-10.0
0.07-0.7
0.5-5.0
Mechanically aided (fan)
1.0-2.0
4.0-8.0
0.07-0.5
0.5-4.0
'Higher L/G reflects those used for gas absorption.
Wet scrubbers remove particles from an exhaust stream by contacting the particles with
liquid, usually water. A number of factors affect particle removal efficiency, including:
• particle-size distribution
• liquid flow rate
• exhaust gas flow rate
• method of contacting
• pressure drop across the scrubber.
As with gaseous pollutant removal (absorption), efficient particle removal requires contact
between the exhaust stream (containing particles) and the scrubbing liquid. However, particle
removal occurs instantaneously upon contact with the liquid, whereas efficient absorption
requires a long contact time. Therefore, efficient particle removal requires high relative
velocities (gas versus liquid velocity).
Estimating Collection Efficiency and Pressure Drop
A number of theories have been developed from basic particle-movement principles to explain
the action of wet scrubbing systems. Many of these start from firm scientific concepts, but
yield only qualitative results when predicting collection efficiencies or pressure drops. The
interaction of particulate matter having a given particle-size distribution with water droplets
having another size distribution is not easy to express in quantitative terms. As a result of this
complexity, experimentally determined parameters are usually needed to approach reality
(Beachler andjahnke 1981).
9-2
-------
Collection Efficiency
Collection efficiency is frequently expressed in terms of penetration. Penetration is defined as
the fraction of particles (in the exhaust stream) that passes through the scrubber uncollected.
Penetration is the opposite of the fraction of particles collected, and is expressed as:
(Eq. 9-1) Pt = 1 - 77
Where: Pt = penetration
7j = collection efficiency.
Wet scrubbers usually have an efficiency curve that fits the relationship of:
(Eq. 9-2) 17 = 1 - e->(1J~~>
Where: 7j = collection efficiency
e = exponential function
f(system) = some function of the scrubbing system variables.
By substituting for efficiency, penetration can be expressed as:
(Eq. 9-3) Pt = 1 — 77
SB 1 — (1 — g-/
—- g-/(lyatm)
In testing the design of a specific scrubber, the vendor can measure operating variables and
the collection efficiency of the unit. These data can then be used to evaluate the efficiency of
the system. An equation for the scrubbing system variables, f(system), can be developed for
that particular design. Scrubber vendors and various consultants have, in fact, developed
equations and assembled data that can be used to design and evaluate their specific scrubbers.
Unfortunately, much of this information is proprietary. In addition, an equation that has
been designed for a venturi scrubber may not work well for evaluating the design of an orifice
or cyclonic scrubber. In other words, there is not one specific equation that can be used to
estimate the collection efficiency of every scrubbing system. A number of equations used for
predicting collection efficiency can be found in the Wet Scrubber System Study (Calvert et al.
1972).
Model for Estimating Venturi Scrubber Efficiency
One method used to predict particle collection efficiency in a venturi scrubber is called the
infinite-throat model (Yung et al. 1977). This model is a refined version of the Calvert cor-
relation given in the Wet Scrubber System Study, Volume I, Scrubber Handbook (Calvert et
al. 1972). The equations presented in the infinite-throat model assume that all panicles are
captured by the water in the throat section of the venturi. Two studies found that this method
correlated very well with actual scrubber operating data (Yung et al. 1977 and Calvert et al.
1978).
9-3
-------
The equations listed in the model can be used to predict the penetration (Pt) for one par-
ticle size (diameter). To get an overall penetration (Pt), you must integrate over the entire
particle-size distribution. Equation 9-4 (penetration for one panicle si2e) was solved for the
overall penetration assuming a log-normal particle-size distribution. These results are plotted
in Figures 9-1 through 9-3 (Yung et al. 1977). In these figures, Pt, overall penetration, is
plotted versus B, a dimensionless parameter characterizing the liquid-to-gas ratio, with KPS, a
dimensionless inertial parameter for mass-median diameter. Each figure has been plotted for a
different geometric standard deviation, i.e., 2.5, 5.0, and 7.5. By knowing the panicle-size
distribution of the dust from an industrial source and the operating conditions of the scrub-
ber, the collection efficiency (penetration) can be estimated using Figures 9-1, 9-2, or 9-3.
Infinite-Throat Model for Predicting Venturi Scrubber Performance
(using metric units)
Equation number
Equation
Eq. 9-4
Eq. 9-5
Nukiyama Tanasawa
equation
41^+ 4.2-5.02 K™ 1 +
lnPt(dp)= - B
Where:
0-7 \ /%:
kJ tan V 0.7
K^ + 0.7
Pt (d,,)= penetration for one panicle size
B = parameter characterizing the liquid-to-gas ratio,
dimensionless
Kpo = inertial parameter at throat entrance
dimensionless
Note: Equation 9-4 was developed assuming that the venturi has an infinite-
sized throat length. This is valid only when I is greater than 2.0.
3 2, Co Qi
(=¦
2d,Q,
Where:
(= throat length parameter, dimensionless
I, — venturi throat length, cm
C0 = drag coefficient for the liquid at the throat
entrance, dimensionless
q, = gas density, g/cm3
d* = droplet diameter, cm
q, = liquid density, g/cms
50
dd= — + 91.8(L/G)1 s
v*»
Where: db = droplet diameter, cm
= gas velocity in the throat, cm/s
L/G = liquid-to-gas ratio, dimensionless
9-4
-------
Infinite-Throat Model for Predicting Venturi Scrubber Performance
(continued)
Equation number
Equation
Eq. 9-6
B = (L/G) -IL.
QtCD
Where: B = parameter characterizing liquid-to-gas ratio,
dimensionless
L/G= liquid-to-gas ratio, dimensionless
Qi = liquid density, kg/ms
qm = gas density, kg/m3
CD = drag coefficient for the liquid at the throat
entrance, dimensionless
Eq. 9-7
9^d,
Where: Kp„ = inertial parameter at the throat entrance,
dimensionless
dp = particle aerodynamic resistance diameter, cmA
v„ = gas velocity in the throat, cm/s
lit = gas viscosity, g/cm*s
d,f = droplet diameter, cm
Eq. 9-8
jr ^p«Iv*»
Where: Kp, = inertial parameter for mass-median diameter,
dimensionless
dpf = particle aerodynamic geometric mean
diameter, cmA
vp = gas velocity in the throat, cm/s
lit = gas viscosity, g/cm*s
d,f = droplet diameter, cm
Eq. 9-9
24
C0 = 0.22 -t- —— (1 + 0.15 N^)
Where: CD = drag coefficient for the liquid at the throat
entrance, dimensionless
Nr«, = Reynolds Number for the liquid droplet at
the throat inlet, dimensionless
9-5
-------
Infinite-Throat Model for Predicting Venturi Scrubber Performance
(continued)
Equation number
Equation
Eq. 9-10
Np„= V"4'
V,
Where: = Reynolds Number for the liquid at the
throat entrance, dimensionless
v„ = gas velocity in the throat, cm/s
v, = gas kinematic viscosity, cmVs
cU = droplet diameter, cm
Eq. 9-11
dp, = dp,(Qx Cp)0'5
Where: dp, = particle aerodynamic geometric mean
diameter, paaA
dp, = particle physical, or Stokes, diameter, fim
Cf= Cunningham slip correction factor, dimensionless
Cp = particle density, g/cm3
Eq. 9-12
+ (6.21 x 10-.)T
d„
Where: Q= Cunningham slip correction factor,
dimensionless
T = absolute temperature, K
dp, = particle physical, or Stokes, diameter, /xm
Source: Yung et al. 1977.
9-6
-------
c
o
V
c
XI
GU
>
o
0.05
2000-=
10,000
0.01
0.005
v
V
E
'-3
c
-3
V
a ¥
e-s
v
E
«
u
ret
cu
e
u
e
0.001
0.5 1 2 5
B, parameter for liquid-to-gas ratio
Figure 9-1. Overall penetration, Ft, versus B
with K„ as a parameter, where the
geometric standard deviation, a,m, is
equal to 2.5.
Source: Yung et al. 1977.
c
£L
>
©
5000
10,000
1000"
0.05
&
£
g
-5
c
2
v
in ^r
CO aj
Ec
>- .2
Q <«
w*
u
2 S
a.
0.01
0.5 1 2
B, parameter for liquid-to-gas ratio
Figure 9-2. Overall penetration, Ft, versus B
with K„ as a parameter, where the
geometric standard deviation, a,m,
is equal to 5.0.
V
c
V
a.
>
O
= 10,000
= 20.000
E
£
M
C
.2
0J
E
S
E c
" .2
V
5
0.05
0.01
0.5 1 2 5
B, parameter for liquid-to-gas ratio
Figure 9-3. Overall penetration, Pt, versus B
with K,,, as a parameter, where the
geometric standard deviation, a,m,
is equal to 7.5.
9-7
-------
Example 9-1 will illustrate how to use the infinite-throat model to predict the performance
of a venturi scrubber. When using the equations given in the model, make sure that the units
for each equation are consistent.
Example 9-1
Cheeps Disposal Inc. is planning to install a hazardous-waste incinerator that will burn both
liquid and solid waste materials. The exhaust gas from the incinerator will pass through a
quench spray and then into a venturi scrubber. Caustic will be added to the scrubbing liquor
to remove any HC1 from the flue gas and to control the pH of the scrubbing liquor. The
uncontrolled particulate emissions leaving the incinerator are estimated to be 1100 kg/h (max-
imum average). The local air pollution regulation states that particulate emissions must not
exceed 10 kg/h. Using the following data, estimate the overall collection efficiency of the
scrubbing system.
Mass-median particle size (physical), dp, = 9.0 /zm
Geometric standard deviation, a,m = 2.5
Particle density, qp= 1.9 g/cm3
Gas viscosity, /x, = 2.0 x 10"4 g/cm«s
Gas kinematic viscosity, vt = 0.2 craVs
Gas density, g,= 1.0 kg/m3
Gas flow rate, Q,g = 15 m3/s
Gas velocity in venturi throat, vp = 9000 cm/s
Gas temperature (in venturi), Tt = 80oC
Water temperature, T, = 30°C
Liquid density, q,= 1000 kg/m3
Liquid flow rate, = 0.014 m3/s
Liquid-to-gas ratio, L/G= 0.0009 L/m3
9-8
-------
Solution
1. The mass-median particle size (physical), dpJ, is
9.0 /Am. Since the particle aerodynamic
geometric mean diameter, dp„ is not known,
you must use Equation 9-11 to calculate dp,
and Equation 9-12 to calculate the
Cunningham slip correction factor, C/.
_ , (6.21 x 10~4)T
(Eq. 9-12) Q= 1 +
(Eq. 9-11) dpl = dp, (C/ X gp)0 5
Note: This step would not have been required if the
particle diameter had been given as the aero-
dynamic geometric mean diameter, dp,, and
expressed in units of /imA.
C/= 1 +
= 1.024
(6.21 x lQ-4)(273 + 80)
9
dp, = 9 /tm(1.024 x 1.9 g/cm3)0 5
= 12.6 fimA
= 12.6 x 10"4 cmA
2. Calculate the droplet diameter, cU, from Equa-
tion 9-5 (Nukiyama Tanasawa equation).
(Eq. 9-5) cL= — + 91.8(L/G)1S
v„
-------
7. The geometric standard deviation, agm, is 2.5.
The overall penetration, Pt, can be found
from Figure 9-4. atm =2.5
B= 1.43
KPf = 992
Read Pt = 0.008
c
o
o
e
V
a.
v
>
o
0.01
0.008
0.005
c
*5
V
e
TO
S c
u o
.3 '3
¦— c
u U
i! £
o
c .^3
a.
0.001
0.5 1 1.43 2 5
B, parameter for liquid-co-gas ratio
Source: Yung et al. 1977.
Figure 9-4. Overall penetration, Pt, for
Example 9-1, where the standard
deviation, a,m, is equal to 2.5.
8. The collection efficiency can be calculated
using the equation below.
T?=l-Pt 77 = 1 — 0.008
= 0.992
= 99.2%
9-10
-------
9. The local regulations state that the particulate
emissions cannot exceed 10 kg/h. The required
collection efficiency can be calculated by using
the equation below.
required
Where:
dust,* - dustoU
dust,„
dust,, = dust concentration leading
into the venturi
dust**, = dust concentration leaving
the venturi
required '
1100 kg/h - 10 kg/h
1100 kg/h
= 0.991
= 99.1%
The estimated efficiency of the venturi scrub-
ber is slightly higher than the required
efficiency.
Note: Figures 9-1 through 9-3 can also be used to determine some of the required operating variables. This
can be done by solving the example problem in reverse. By entering the figures at the required effi-
ciency (or Pt). one can obtain various sets of Kpf and B values. These values for Kpf and B can be used
to calculate the required L/G ratio or v„ for a specific collection efficiency.
Cut Power Method
One empirical correlation that has been used to predict the collection efficiency of a scrubber
is the cut power method. In this method, developed by Calvert, penetration is a function of
the cut diameter of the particles to be collected by the scrubber. The cut diameter is the
diameter of the particles that are collected by the scrubber with at least 50% efficiency. Since
scrubbers have limits to the size of particles they can collect, knowledge of the cut diameter is
useful in evaluating the scrubbing system.
In the cut power method, penetration is a function of the particle diameter and is given as:
(Eq. 9-13)
Where: Pt = penetration
A^, = parameter characterizing the particle-size distribution
Bcut = empirically determined constant, depending on the scrubber design
dp = aerodynamic diameter of the particle.
Penetration, calculated by Equation 9-13, is given for only one particle size (d„). To obtain
the overall penetration, the equation can be integrated over the log-normal panicle-size
distribution. Calvert has developed a method of determining the cut diameter required to
achieve a given collection efficiency. By mathematically integrating Pt over a log-normal
distribution of particles and by varying the geometric standard deviation, as„, and the
geometric mean particle diameter, dPJ, the overall penetration, Pt, can be obtained. Figure
9-5 plots the overall penetration as a function of the required cut diameter, (dp)cur (Calvert
1972).
9-11
-------
Figure 9-5 was developed using Equation 9-13, Pt = e'Acurdp where B = 2. For plate
towers, B = 2, but is only ~2 for Venturis under certain conditions. For centrifugal scrubbers,
B = 0.7 and, therefore, Figure 9-5 should not be used as is. Further limitations and models
developed for specific devices using the cut power method are discussed by Calvert (1972).
The application of these models to actual operating systems has not been documented ade-
quately in the open literature.
Example 9-2 will illustrate how to use the cut power method to estimate the cut diameter
for a venturi scrubber.
Example 9-2
Given similar conditions as in Example 9-1, estimate the cut diameter for a venturi scrubber.
The data below are approximate.
Geometric standard deviation, crgm = 2.5
Particle aerodynamic geometric mean diameter, <^,= 12.6 /onA
Required efficiency, 77 = 99.1% or 0.991
Solution
1. For an efficiency of 99.1%, the overall pene-
tration can be calculated from Pt = 1 — 77.
Pt = 1 — 0.991
= 0.009
2. The overall penetration is 0.009, and the
geometric standard deviation is 2.5. Using
Figure 9-5, read (dp)™,/dp,.
Pt =0.009
atm = 2.5
(dpWdp.~0.09
0.5
0.1
0.01
0.001
0.001
0.01
0.1
1.0
(dp)cw*/dpg
Source: Calvert 1977.
Figure 9-5. Penetration versus cut diameter.
9-12
-------
3. The cut diameter, (dp)cu,, is calculated from
=0.09
dPt
(dp)e«r = (0.09) dp,.
From the information presented in Lesson 2, a
venturi scrubber is capable of operating with a
cut diameter of 1.13 /ixnA. However, the
pressure drop and other operating conditions
of the scrubber must be maintained to achieve
a high collection efficiency.
Contact Power Theory
A more general theory for estimating collection efficiency is the contact power theory. This
theory is based on a series of experimental observations made by Lapple and Kamack (1955).
The fundamental assumption of the theory is:
"When compared at the same power consumption, all scrubbers give substantially
the same degree of collection of a given dispersed dust, regardless of the mechanism
involved and regardless of whether the pressure drop is obtained by high gas flow
rates or high water flow rates." (Lapple and Kamack 1955)
In other words, collection efficiency is a function of how much power the scrubber uses,
and not of how the scrubber is designed. This has a number of implications in the evaluation
and selection of wet collectors. Once it is realized that a certain amount of power is needed
for a required collection efficiency, the claims about specially located nozzles, baffles, etc. can
be evaluated more objectively. The choice between two different scrubbers with the same
power requirements may depend primarily on ease of maintenance.
Semrau (1959 and 1963) developed the contact power theory from the work of Lapple and
Kamack (1955). The theory, as developed by Semrau, is empirical in approach and relates the
total pressure loss, PT, of the system to the collection efficiency.
The total pressure loss is expressed in terms of the power expended to inject the liquid into
the scrubber plus the power needed to move the process gas through the system.
(Eq. 9-14) pT=pG + pL
Where: PT= total contacting power, kWh/1000 m3 (hp/1000 acfm)
Pc = power input from gas stream, kWh/1000 m3 (hp/1000 acfm)
Pl — power input from liquid injection, kWh/1000 m3 (hp/1000 acfm)
[Note: The total pressure loss, PT, should not be confused with penetration, Pt, defined in the
previous section. Penetration is the symbol used by Calvert to express the fraction of par-
ticulate matter escaping from a collector.]
(dp)cut = (0.09)(12.6 iimA)
= 1.13 umA
9-13
-------
The power expended in moving the gas through the system, Pa, is expressed in terms of the
scrubber pressure drop.
(Eq. 9-15) Pc = 2.724x 10~4 Ap, kWh/1000 m3 (metric units)
or
Pa — 0.1575 Ap, hp/1000 acfm (English units)
Where: Ap = pressure drop, kPa (in. H20)
The power expended in the liquid stream, PL, is expressed as:
(Eq. 9-16) P£ = 0.28 pt (Q^/Q.c), kWh/1000 ms (metric units)
or
PL = 0.583 pL (Qj./Q,g), hp/1000 acfm (English units)
Where: p£ = liquid inlet pressure, 100 kPa (lb/in.1)
Qt = liquid feed rate, m3/h (gal/min)
Qc = gas flow rate, mVh (ft3/min).
The constants given in the expressions for Pc and PL incorporate conversion factors to put the
terms on a consistent basis.
The total power can therefore be expressed as:
(Eq. 9-17) Pt=Pg + Pl
= 2.724x lO"" Ap + 0.28 pL (Qj./Q.a), kWh/1000 m3
or
= 0.1575 Ap +0.583 pt (Q^/Qc), hp/1000 acfm.
The problem now is to correlate this with scrubber efficiency.
Equation 9-2 of this lesson shows that efficiency is an exponential function of the system
variables for most types of collectors.
(Eq. 9-18)
Semrau defines the function of the system variables, f(system), as:
(Eq. 9-19) f(system) = Nr = aPTa
Where: Nt = number of transfer units
Pt= total contacting power
a and 0 = empirical constants which are determined from experiment and depend
on the characteristics of the panicles.
The efficiency then becomes:
(Eq. 9-20) r7=l_e-^ra.
Table 9-2 gives values of a and /3 for different industries. The values of a and j3 can be
used in either the metric or English units.
9-14
-------
Table 9-2. Parameters a. and 0 for the contact power theory.
Scrubber design
Aerosol
a
&
Venturi
Talc dust
2.97
0.362
Phosphoric acid mist
1.33
0.647
Foundry cupola dust
1.35
0.621
Open-hearth steel furnace fume
1.26
0.569
Odorous mist
0.363
1.41
Venturi evaporator
Hot black liquor gas
0.522
0.861
Venturi and cyclonic spray
Lime kiln dust (raw)
1.47
1.05
Black liquor furnace fume
1.75
0.620
Ferrosilicon furnace fume
0.870
0.459
Lime kiln dust (prewashed)
0.915
1.05
Black liquor fume
0.740
0.861
Venturi condensation
scrubber with:
1. Mechanical spray
Copper sulfate
0.390
1.14
generation
2. Hydraulic nozzles
Copper sulfate
0.562
1.06
Orifice
Talc dust
2.70
0.362
Cyclone
Talc dust
1.16
0.655
Source: Semrau 1960.
The contact power theory cannot predict efficiency from a given panicle-size distribution as
can the cut power method. The contact power theory gives a relationship which is indepen-
dent of the size of the scrubber. With this observation, a small pilot scrubber could first be
used to determine the pressure drop needed for the required collection efficiency. The full-
scale scrubber design could then be scaled up from the pilot information.
Example 9-3
A wet scrubber is to be used to control particulate emissions from a foundry cupola. Stack test
results reveal that the particulate emissions must be reduced by 85% to meet emission stan-
dards. If a 100-acfm pilot unit is operated with a water flow rate of 0.5 gal/min at a water
pressure of 80 psi, what pressure drop (Ap) would be needed across a 10,000-acfm scrubber
unit?
Solution
1. From Table 9-2, read the a and j3 parameters
for foundry cupola dust. a = 1.35
0 = 0.621
2. Calculate the number of transfer units, Nt,
using Equation 9-18.
(Eq. 9-18) 77= 1 -e-"< N, = In
N, = ln = In 6.66
1~v =1.896
9-15
-------
3. Calculate the total contacting power, PT, using
Equation 9-19.
(Eq. 9-19) Nt = aPra
1.896=1.35 Pr0 821
1.896
= Pt'
0-621
1.35
1.404 = Pr° 421
In 1.404 = 0.621 lnP7
0.3393 = 0.621 lnPr
0.5464 = lnPr
PT= 1.73 hp/1000 acfm
4. Calculate the pressure drop, Ap, using Equa-
tion 9-17.
(Eq. 9-17) PT= 0.1575 Ap +0.583 pt ( ^
PT= 0.1575 Ap + 0.583 pt
1.73 = 0.1575 Ap + 0.583(80)
Ap = 9.5 in, H20
0.5
Too
From the data in Table 9-2, you can see that the usefulness of Equation 9-20 is limited due
to the lack of a and 0 values. However, the concept of the contact power theory is still a very
useful tool in evaluating scrubber design. Since the theory does correlate well with operating
data, and it is independent of scrubber size, the theory has applications in scaling up designs
from pilot plant data. In addition, the basic principle of the contact power theory can be
applied to specific sources of interest. For example, in a regulatory analysis of wet scrubbing
systems for coal-fired utility boilers, Figure 9-6 was developed using the contact power theory
(Kashdan 1979). This figure plots power consumption versus outlet dust loading (instead of
transfer units) for the operating points of 12 utility boilers. The curve fits the equation
y = 0.68x"1-41, where y = outlet grain loading and x = theoretical power consumption calculated
using Equation 9-17. The good fit is quite remarkable given the variety of coals, boilers, proc-
ess variables, inlet panicle-size distribution, and scrubber designs among the different plants
(Kashdan 1979).
The concept of the contact power theory does have limitations. It does not apply to a
number of new wet collecting systems where a combination of collecting mechanisms are used,
such as condensation scrubbers. Also, the theory applies best when the power is applied in one
scrubbing area (Mcllvaine 1977), such as in a venturi scrubber. Multiple-staged devices and
packed towers will have collection efficiencies varying from those of a venturi scrubber for a
given power input.
9-16
-------
0.08
0.07
o
i i r™
1 1 II l_
o
in
-a
0.06
0.05
—
&
0.04
o\
— AO
0
be
*5
C9
£
0.03
— o
"
-------
Using Equation 9-21 for the conditions given in Example 9-1, we get:
Vgt = 9000 cm/s
L/G = 0.0009 L/m3
Ap = 8.24 x 10~4(9000)2(0.0009)
= 60 cm H20
Using Pilot Methods to Design Scrubbers
The semi-empirical theories previously discussed are useful for scrubber design and evaluation
exercises because they can give qualitatively correct information. However, they have a
number of practical limitations. It is not common practice to choose scrubber systems based
only on this information. The uncertainties involved in particle-size determinations and the
questions associated with using empirically determined parameters restrict the use of
theoretical methods. Basically, too many variables are involved and accounting for them all in
a simple theory is too difficult. The time and expense needed to obtain good input data for
these methods may be better spent in developing pilot plant information.
Scrubbers that work primarily through impaction mechanisms have certain performance
characteristics (such as efficiency and pressure drop) which are independent of scale. This
consequence of the contact power principle provides the basis for using pilot systems. By using
a small-scale scrubber (100 to 1000 cfm) on the exhaust gas stream, the effectiveness of the
equipment for removing the actual panicles in the gas can be experimentally determined.
Pilot systems ranging from 170 m3/h (100 cfm) units to one-tenth the size of full-scale
plants have been developed in the past. Mcllvaine (1977) has compared the effectiveness of
the various design methods. His work is summarized in Table 9-3.
Table 9-3. Methods for predicting venturi scrubber pressure requirements.
Description
Expense
(relative scale)
Time
(months)
Most reliable
1/10 size full-scale plants
100-1000
12-24
2000-cfm pilot units
30
3-6
100-cfm pilot units
5
2-3
Empirical curves based on
0.2
0.2
similar processes
Impactor in situ particle
2
1
'
' sizing
Least reliable
The design of a wet collector system for a particulate-emission problem requires more than
the application of a few design equations. The experience of scrubber manufacturers with
specific industry installations, coupled with the use of pilot units, provides more reliable ways
to determine the size of a system for a wide range of operating conditions. In many cases,
theoretical models can complement such studies and provide qualitative data for wet collector
evaluations.
9-18
-------
Review of Design Criteria for Permits
The principal design criteria are the exhaust flow to the scrubber, measured in units of
mVmin (ft3/min, or acfm), and the dust concentration, measured in units of g/m3 (lb/ft3, or
gr/ft3). The exhaust volume and dust loading are set by the process exhaust conditions. Once
these criteria are known, the vendor can begin to design the scrubber for the specific applica-
tion. A thorough review of the design plans should consider the factors presented below.
Dust properties—type, shape, density, and size of the dust particles; average and maximum
concentrations in the process exhaust stream. If the scrubber is to be installed on an
existing source, a stack test to measure dust concentration and particle-size distribution (cut
diameter and standard deviation) should be performed. If the scrubber is installed on a new
source, these data could be obtained from a similar installation.
Exhaust gas characteristics—average and maximum exhaust flow rates to the scrubber;
chemical properties such as dew point, corrosiveness, pH, and solubility of the pollutants to
be removed should be measured or accurately estimated.
Liquid flow—the type of scrubbing liquid and the rate at which the liquid is supplied to
the scrubber. If the scrubbing liquid is to be recirculated, the pH and amount of suspended
solids should be monitored to ensure continuous reliability of the scrubbing system.
Pressure drop—the pressure drop (gas-side) at which the scrubber will operate; the scrub-
ber design should also include a means for monitoring the pressure drop across the system,
usually by manometers.
Removal of entrained liquid—mists and liquid droplets that become entrained in the
"scrubbed" exhaust stream should be removed before exiting the stack. Some type of
entrainment separator, or mist eliminator, should be included in the design.
Emission requirements—collection efficiency in terms of grain loading and opacity regula-
tions for particulate matter and concentration regulations for gaseous pollutants; collection
efficiency can be high (95 to 99%) if the scrubber is properly designed. The agency review
engineer can use the equations given in this lesson to estimate the scrubber efficiency.
However, these equations can only predict the general collection efficiency of the system,
and they should not be used as the basis to either accept or reject the design plans sub-
mitted for the permit process.
Summary
When checking the design plans for the permit process, the agency engineer should check its
files or another agency's files for similar applications for scrubber installations. A review of
these data will help determine if the scrubber design specifications submitted by the industrial
source's officials are adequate to achieve particle removal efficiency for compliance with the
regulations. The agency engineer should require the source owner/operator to conduct stack
tests (once the source is operating) to determine if the source is in compliance with local,
State, and Federal regulations. The agency engineer should also require that the source
owner/operator submit an operation and maintenance schedule that will help keep the scrub-
ber system on line.
9-19
-------
Review Exercise
1. The scrubber used most often to remove particulate matter
frnm pvhansf ctr-pams is a scrubber.
2. True or False? Efficient particle removal requires low
gas-to-liquid (relative) velocities. ~y~
1. venturi
3. The term penetration is defined as
a. the fraction of particles collected in a scrubber.
b. the amount of gaseous pollutants absorbed in the scrub-
bing liquor.
c. the fraction of particles that passes through a scrubber
uncollected.
2. False
4. The cut power method is an empirical correlation used to
predict the penetration. The penetration is a function of
a. the cut diameter of the particles that are collected by
the scrubber. r ^
b. the cut-throat velocity of the scrubber.
c. the amount of power that is supplied to the scrubber.
d. the pressure drop across the scrubber. -j/
3. c. the fraction of par-
ticles that passes
through a scrubber
uncollected.
5. Cut diameter is
a. the cut-off size of the particles that are not collected.
b. the diameter of the particles that are collected with at
least 100% efficiency.
c. the diameter of the particles that are collected with at
least 50% efficiency. (_y
4. a. the cut diameter of
the particles that are
collected by the
scrubber.
6. In the equation used in the contact power theory,
Pt=Pg + Pl, the symbol PT represents
a. the penetration of the system. ^
b. the collection efficiency.
c. the total pressure loss, or contacting power, of the
scrubbing system.
5. c. the diameter of the
particles that are
collected with at least
50% efficiency.
' i
7. According to the contact power theory, the lower/higher'
the pressure drop is across the scrubbing system, the
higher the collection efficiency will be.
6. c. the total pressure loss,
or contacting power,
of the scrubbing
system.
8. Which of the following factors affect the pressure drop of
a scrubbing system?
a. scrubber design and geometry
b. gas velocity ry*.
c. liquid-to-gas ratio
d. all of the above
7. higher
8. d. all of the above
9-20
-------
References
Beachler, D. S., andjahnke, J. A. October 1981. Control of particulate emissions. APTI
Course 413, EPA 450/2-80-006. U.S. Environmental Protection Agency. Research Triangle
Park, NC.
Brady, J. D., and Legatski, L. K. 1977. Venturi scrubbers. In Air pollution control and
design handbook. P. N. Cheremisinoff and R. A. Young, eds. New York: Marcel Dekker,
Inc.
Calvert, S. 1977. How to choose a particulate scrubber. Chem. Eng. 84:54-68.
Calvert, S.; Barbarika, H. F.; and Monahn, G. M. 1978. Evaluation of three industrial
particulate scrubbers. EPA 600/2-78-032. U.S. Environmental Protection Agency.
Cincinnati, OH.
Calvert, S.; Goldshmid, J.; Leith, D.; and Mehta, D. 1972. Wet scrubber system study, vol-
ume I: scrubber handbook. EPA-R2-72-118a. U.S. Environmental Protection Agency.
Research Triangle Park, NC.
Environmental Protection Agency. 1982. Control techinques for particulate emissions from
stationary sources—volume I. EPA 450/3-81-005a. Research Triangle Park, NC.
Environmental Protection Agency. 1973. Air pollution engineering manual. AP40. Research
Triangle Park, NC.
Kashdan, E. R., and Ranade, M. B. 1979. Design guidelines for an optimum scrubber system.
EPA 600/7-79-018. U.S. Environmental Protection Agency. Research Triangle Park, NC.
Lapple, C. E., and Kamack, H. J. 1955. Performance of wet dust scrubbers. Chem. Eng.
Prog. 51:110-121.
Mcllvaine, R. W. June 1977. When to pilot and when to use theoretical predictions of
required venturi pressure drop. Paper presented at meeting of the Air Pollution Control
Association, at Toronto, Canada.
Nukiyama, S., and Tanasawa, Y. 1983. An experiment on atomization of liquid by means
of air stream (in Japanese). Trans. Soc. Mech. Eng. Japan 4:86.
Perry, J. H., ed. 1973. Chemical engineers' handbook, 5th ed. New York: McGraw-Hill
Book Co.
Rimberg, D., and Peng, Y. M. 1977. Aerosol collection by falling droplets. In Air pollution
control and design handbook, pp. 747-777. P. N. Cheremisinoff and R. A. Young, eds.
New York: Marcel Dekker, Inc.
Rimberg, D. B. March 1979. Tips and techniques on air pollution control equipment O & M.
Pollut. Eng., pp. 32-35.
Semrau, K. T. 1977. Practical process design of particulate scrubbers. Chem. Eng. 84:87-91.
Semrau, K. T. 1963. Dust scrubber design—a critique on the state of the art./. Air Poll.
Control Assoc. 13:587-593.
Semrau, K. T. 1960. Correlation of dust scrubber efficiency./. Air Poll. Control Assoc.
10:200-207.
9-21
-------
Sparks, L. E. 1978. SR-52 programmable calculator programs for venturi scrubbers and
electrostatic precipitators. EPA 600/7-78-026. U.S. Environmental Protection Agency.
Research Triangle Park, NC.
Stairmand, C. J. 1956. The design and performance of modern gas-cleaning equipment.
J. Inst. Fuel 29:58-81.
Strauss, W. 1975. Industrial gas cleaning. Oxford: Pergamon Press.
Yung, S.; Calvert, S.; and Barbarika, H. F. 1977. Venturi scrubber performance model.
EPA 600/2-77-172. U.S. Environmental Protection Agency. Cincinnati, OH.
9-22
-------
Lesson 10
Design Review of Absorbers Used
for Gaseous Pollutants
Lesson Goal and Objectives
Goal
To familiarize you with the factors to be considered when reviewing absorber design plans for
the permit process.
Objectives
Upon completing this lesson, you should be able to —
1. recall at least four scrubber design factors,
2. estimate the liquid flow rate, the diameter, and the packing height of a packed tower
using appropriate tables and equations, and
3. estimate the number of plates and the height of a plate tower using appropriate tables
and equations.
Introduction
The design of an absorber vised to reduce gaseous pollutants from process exhaust streams
involves many factors including the pollutant collection efficiency, pollutant solubility in the
absorbing liquid, liquid-co-gas ratio, exhaust flow rate, pressure drop, and many construction
details of the absorbers such as packing, plates, liquid distributors, entrainment separators,
and corrosion-resistant materials. These have been discussed in detail in the previous lessons.
Air pollution control agency officers who review design plans for absorbers should consider
these factors during the review process.
The previous lessons stated that all wet scrubbing systems are able to collect particulate and
gaseous pollutants emitted from process exhaust streams. However, spray towers, plate towers,
packed towers, and moving-bed scrubbers are generally used to reduce gaseous pollutants.
This lesson will focus on equations used to estimate liquid flow rate, the diameter and the
height of a packed tower, and the diameter and number of plates used in a plate tower to
achieve a specified pollucant removal efficiency.
10-1
-------
Absorbers remove gaseous pollutants by dissolving them into a liquid called the absorbent.
In designing absorbers, optimum absorption efficiency can be achieved by:
• providing a large interfacial contact area,
• providing for good mixing between the gas and liquid phases,
• allowing sufficient residence, or contact, time between the phases, and
• choosing a liquid in which the gaseous pollutant is very soluble.
Absorption
Absorption refers to the transfer of a gaseous pollutant from a gas phase to a liquid phase.
More specifically, in air pollution control, absorption involves the removal of objectionable
gaseous pollutants from a process stream by dissolving them in a liquid.
Some common terms used when discussing the absorption process follow:
Absorbent—the liquid, usually water, into which the pollutant is absorbed.
Solute, or absorbate—the gaseous pollutant being absorbed, such as S02, H,S, etc.
Carrier gas—the inert portion of the gas stream, usually air, from which the pollutant is to
be removed.
Interface—the area where the gas phase and the absorbent contact each other.
Solubility—the capability of a gas to be dissolved in a liquid.
Absorption is a mass-transfer operation. In absorption, mass transfer of the gaseous pollu-
tant into the liquid occurs as a result of a concentration difference (of the pollutant) between
the liquid and gas phases. Absorption continues as long as a concentration difference exists
where the gaseous pollutant and liquid are not in equilibrium with each other. The concen-
tration difference depends on the solubility of the gaseous pollutant into the liquid.
Solubility
A very important factor affecting the amount of a pollutant, or solute, that can be absorbed
is its solubility. Solubility is a function of both the temperature and, to a lesser extent, the
pressure of the system. As temperature increases, the amount of gas that can be absorbed by a
liquid decreases. From the ideal gas law: as temperature increases, the volume of a gas also
increases; therefore, at the higher temperature, less gas is absorbed due to the increased
volume it occupies. Pressure affects the solubility of a gas in the opposite manner. By increas-
ing the pressure of a system, the amount of gas absorbed generally increases.
The solubility of a specific gas in a given liquid is defined at a designated temperature and
pressure. Table 10-1 presents data on the solubility of S02 gas in water at 101 kPa, or 1 atm,
and various temperatures. In determining solubility data, the partial pressure (in mm Hg) is
measured with the concentration (in grams of solute per 100 grams of liquid) of the solute in
the liquid. The data in Table 10-1 were taken from The International Critical Tables, a good
source of information concerning gas-liquid systems.
Solubility data are obtained at equilibrium conditions. This involves putting measured
amounts of a gas and a liquid into a closed vessel and allowing it to sit for a period of time.
Eventually, the amount of gas absorbed into the liquid will equal the amount coming out of
the solution. At this point, there is no net transfer of mass to either phase, and the concentra-
tion of the gas in both the gaseous and liquid phases remains constant. The gas-liquid system
is at equilibrium.
10-2
-------
Table 10-1. Partial pressure of SO» in aqueous solution, mm Hg.
Grains of
SO, per
10°C
20 °C
30°C
40 °C
50 °C
60 °C
70 °C
100 g H,0
0.0
0.5
21
29
42
60
83
111
144
1.0
42
59
85
120
164
217
281
1.5
64
90
129
181
247
328
426
2.0
86
123
176
245
333
444
581
2.5
108
157
224
311
421
562
739
3.0
130
191
273
378
511
682
897
3.5
153
227
324
447
603
804
—
4.0
176
264
376
518
698
—
—
4.5
199
300
428
588
793
—
—
5.0
223
338
482
661
—
—
-
Equilibrium conditions are important in operating an absorption tower. If equilibrium were
to be reached in the actual operation of an absorption tower, the collection efficiency would
fall to zero at that point since no net mass transfer could occur. The equilibrium concentra-
tion, therefore, limits the amount of solute that can be removed by absorption. The most
common method of analyzing solubility data is to use an equilibrium diagram. An equili-
brium diagram is a plot of the mole fraction of solute in the liquid phase, denoted as x, versus
the mole fraction of solute in the gas phase, denoted as y. Equilibrium lines for the S02 and
water system given in Table 10-1 are plotted in Figure 10-1. Figure 10-1 also illustrates the
temperature dependence of the absorption process. At a constant mole fraction of solute in
the gas (y), the mole fraction of SOj that can be absorbed in the liquid (x) increases as the
temperature decreases.
0.6
u
(Q
C 0.5
o
£ 0.4
o
c
o
0.3
u
a
I 0.2
E
0.014
0.010
! 0.006
x, mole fraction of SOj in water
0.002
Figure 10-1. Equilibrium lines for SO,-HsO systems
at various temperatures.
10-3
-------
Under certain conditions, Henry's law may also be used to express equilibrium solubility of
gas-liquid systems. Henry's law is expressed as:
(Eq. 10-1) p = Hx
Where: p = partial pressure of solute at equilibrium, Pa
x = mole fraction of solute in the liquid, mole fraction
H= Henry's law constant, Pa/mole fraction.
From Equation 10-1 you can see that H has the units of pressure per concentration. Henry's
law can be written in a more useful form by dividing both sides of Equation 10-1 by the total
pressure, Pr, of the system. The left side of the equation becomes the partial pressure divided
by the total pressure, which equals the mole fraction in the gas phase, y. Equation 10-1 now
becomes:
(Eq. 10-2) y = H'x
Where: y = mole fraction of gas in equilibrium with liquid
H' — Henry's law constant, mole fraction in vapor per mole fraction in liquid
x = mole fraction of the solute in the liquid.
Note: H' now depends on the total pressure.
Equation 10-2 is the equation of a straight line, where the slope (m) is equal to H'. Henry's
law can be used to predict solubility only when the equilibrium line is straight. Equilibrium
lines are usually straight when the solute concentrations are very dilute. In air pollution con-
trol applications, this is usually the case. For example, an exhaust stream that contains a
1000-ppm SOj concentration corresponds to a mole fraction of S02 in the gas phase of only
0.001. Figure 10-2 demonstrates that the equilibrium lines are still straight at this low concen-
tration of SOj.
eo
.5
M
o
CO
c
o
u
(9
£
O
0.004 0.008
x, mole fraction of SO* in water
0.012
Figure 10-2. Equilibrium diagram for SO,-HtO system
for the data given in Example 10-1.
10-4
-------
Another restriction on using Henry's law is that it does not hold true for gases that react or
dissociate upon dissolution. If this happens, the gas no longer exists as a simple molecule. For
example, scrubbing HF or HC1 gases with water causes both compounds to dissociate in solu-
tion. In these cases, the equilibrium lines are curved rather than straight. Data on systems
that exhibit curved equilibrium lines must be obtained from experiments.
Henry's law constants for the solubility of several gases in water are listed in Table 10-2.
The units of Henry's law constants are atmospheres per mole fraction. The smaller the con-
stant, the more soluble the gas. Table 10-2 demonstrates that S02 is approximately 100 times
more soluble in water than CO* is in water. The following example illustrates how to develop
an equilibrium diagram from solubility data.
Table 10-2. Henry's law constants for gases in HtO.*
Gas
20 °C
30°C
N,
80.4
92.4
CO
53.6
62.0
H,S
48.3
60.9
0,
40.1
47.5
NO
26.4
31.0
CO*
1.42
1.86
SO,
0.014
0.016
'Expressed in Hx 10"5, atm/mole fraction.
Example 10-1
Given the data in Table 10-3 for the solubility of SOj in pure water at 303 K (30 °C) and
101.3 kPa (760 mm Hg), calculate y and x, plot the equilibrium diagram, and determine if
Henry's law applies.
Table 10-3. Equilibrium data.
CjOj
PjOj
7
X
(g of SO, per 100 g
(partial pressure
(mole fraction of
(mole fraction of
of HxO)
of SO,)
SO, in gas phase)
SO, in liquid phase)
0.5
6 kPa (42 mm Hg)
1.0
11.6 kPa (85 mm Hg)
1.5
18.3 kPa (129 mm Hg)
2.0
24.3 kPa (176 mm Hg)
2.5
30.0 kPa (224 mm Hg)
3.0
36.4 kPa (273 mm Hg)
10-5
-------
Solution
1. The data must first be converted into mole
fraction units. The mole fraction of S02 in the
gas phase, y, is obtained by dividing the par-
tial pressure of S02 by the total pressure of the
system.
P-IOj
~pT
The mole fractions of S02 in the gas phase are
tabulated in Table 10-4.
6 kPa
y=
101.3 kPa
= 0.06
2. The mole fraction of the solute (SO*) in the
liquid phase, x, is obtained by dividing the
moles of S02 dissolved into the solution by the
total moles of liquid.
moles of S02 in solution
x =
moles of S02 in solution + moles of H20
The mole fractions of the solute in the liquid
phase are tabulated in Table 10-4.
moles of S02 in solution = cjo2/64 g SOz
per mole
moles of HjO = 100 g of H20/18 g H20
per mole
= 5.55 moles
x =
Cso,/64
Qo2/64+ 5.55
0J3
64
~ol
77 +5.55
64
= 0.0014
Table 10-4. Equilibrium data for Example 10-1.
g of SO,
Pjo,
(kPa)
y= p/101.3
c«>,/64
Wo*
100 g HtO
Cio,/64 + 5.55
0.5
6.0
0.060
0.0014
1.0
11.6
0.115
0.0028
1.5
18.3
0.180
0.0042
2.0
24.3
0.239
0.0056
2.5
30.0
0.298
0.0070
3.0
36.4
0.359
0.0084
10-6
-------
3. The mole fraction of S02 in air, y, is plotted
against the mole fraction of S02 dissolved in
water, x, in Figure 10-2.
_c
H
o
C/5
0.004 0.008
x, mole fraction of SOi in water
0.012
Figure 10-2. Equilibrium diagram for SOt-HtO system
for the data given in Example 10-1.
The plot in Figure 10-2 is a straight line;
therefore, Henry's law applies. The slope of
the line (Ay/Ax), Henry's law constant (//'), is
approximately equal to 42.7.
Ay
S1°P*= s
0.239-0.180
~ 0.0056-0.0042
= 42.7
10-7
-------
Review Exercise
1. Of the wet collectors listed below, which is/are the best
device(s) for removing gaseous pollutants from process
exhaust streams?
a. packed tower
b. plate tower , 7 /
c. venturi scrubber
d. centrifugal scrubber
e. a and b
2. In the absorption process, the solute is the
a. inert portion of the gas stream.
b. area where the gas phase and liquid phase come into
contact with each other. r
c. gaseous pollutant that is absorbed. (__,
d. capability of a gas to be dissolved in a liquid.
1. e. a and b
3. A very important factor affecting the amount of a pollu-
ranr rhar ran he ahsnrfwl is irs ' • ; t-i.-^v
2. c. gaseous pollutant that
is absorbed.
u
4. In an absorber, as the temperature of the system increases,
the amount of pollutant that can be absorbed
increases/decreases.
3. solubility
V -• '
5. A plot of the mole fraction of the solute in the liquid
phase versus the mole fraction of the solute in the gas
phase is called
a. the partial pressure. kf' ^
b. an equilibrium diagram.
c. a concentration gradient.
4. decreases
6. What is one form of the equation for Henry's law?
a. x = Hp
b. H=xp 1
c. H — x/y
d. y = H'x
5. b. an equilibrium
diagram.
7. In describing the solubility of various gases in water, the
smaller/larger Henry's law constant is, the more soluble
che-'gas is.
6. d. y — H'x
7. smaller
10-8
-------
Absorber Design
Theory
The first step in designing an air pollution control device is to develop a mathematical expres-
sion describing the observed phenomenon. A valid mathematical expression describing
absorber performance makes it possible to determine the proper absorber size for a given set
of conditions, and predict how a change in operating conditions affects absorber performance.
A number of theories, or models, attempt to analytically describe the absorption mechanism.
However, in practice, none of these analytical expressions can solely be used for design
calculations. Experimental or empirical data must also be used to obtain reliable results.
The most widely used model for describing the absorption process is the two-film, or
double-resistance, theory, which was first proposed by Whitman in 1923. The model starts
with the three-step mechanism of absorption previously discussed in Lesson 1. From this
mechanism, the rate of mass transfer was shown to depend on the rate of migration of a
molecule in either the gas or liquid phase. The two-film model starts by assuming that the gas
and liquid phases are in turbulent contact with each other, separated by an interface area
where they meet. This assumption may be correct, but no mathematical expressions ade-
quately describe the transport of a molecule through both phases in turbulent motion.
Therefore, the model proposes that a mass-transfer zone exists to include a small portion
(film) of the gas and liquid phases on either side of the interface. The mass-transfer zone is
comprised of two films, a gas film and a liquid film on their respective sides of the interface.
These films are assumed to flow in a laminar, or streamline, motion. In laminar flow,
molecular motion occurs by diffusion, and can be categorized by mathematical expressions.
This concept of the two-film theory is illustrated in Figure 10-3.
Gas-liquid
Bulk-gas phase interface Bulk-liquid phase
^r-—\
Partial pressure \
I drivine force 1.
1—.—til'
Gas Liquid
film film
Figure 10-3. Visualization of two-film theory.
10-9
-------
According to the two-film theory, for a molecule of substance A to be absorbed, it must
proceed through a series of five steps. The molecule must:
1. migrate from the bulk-gas phase to the gas film,
2. diffuse through the gas film,
3. diffuse across the interface,
4. diffuse through the liquid film, and finally
5. mix into the bulk liquid.
The theory assumes that complete mixing takes place in both gas and liquid bulk phases and
that the interface is at equilibrium with respect to molecules transferring in or out. This
implies that all resistance to movement occurs when the molecule is diffusing through the gas
and liquid films, hence the name double-resistance theory. The partial pressure (concentra-
tion) in the gas phase changes from p^c in the bulk gas to pai at the interface.
A gas concentration is expressed by its partial pressure. Similarly, the concentration in the
liquid changes from cai at the interface to Cal in the bulk phase as mass transfer occurs. The
rate of mass transfer then equals the amount of molecule A transferred multiplied by the
resistance molecule A encounters in diffusing through the films.
Where: NA = rate of transfer of component A, g-mol/h«ml (lb-mole/hr*ft2)
k, = mass-transfer coefficient for gas film, g-mol/h*mi»Pa (lb-mole/hr*ft3»atm)
k, = mass-transfer coefficient for liquid film, g-moI/h*m**Pa (Ib-moie/hr*ft2*atm)
p^c = partial pressure of solute A in the gas
pAi = partial pressure of solute A at the interface
caj = concentration of solute A at the interface
cal — concentration of solute A in the liquid
The mass-transfer coefficients, k, and k, represent the flow resistance the solute encounters
in diffusing through each film respectively (Figure 10-4). An analogy is the resistance electric-
ity encounters as it flows through a circuit.
Equations 10-3 and 10-4 define the general case of absorption and are applicable to both
curved and straight equilibrium lines. In practice, Equations 10-3 and 10-4 are difficult to
use, since it is impossible to measure the interface concentrations, pai and Cai- The interface is
a fictitious state used in the model to represent an observed phenomenon. The interface con-
centrations can be avoided by defining the mass-transfer system at equilibrium conditions and
combining the individual film resistances into an overall resistance. If the equilibrium line is
straight, the rate of absorption is given by the equations below.
Where: N* = rate of transfer of component A, g-mol/h«m2 (lb-mole/hr*ft2)
pjj = equilibrium partial pressure of solute A at operating conditions
c* = equilibrium concentration of solute A at operating conditions
Kog = overall mass-transfer coefficient based on gas phase, g-mol/h«m2«Pa
(lb-mole/hr*ft2*atm)
K<>£ = overall mass-transfer coefficient based on liquid phase, g-mol/h«m2«Pa
(lb-mole/hr*ft2*atm)
pac = partial pressure of solute A in the gas
cal = concentration of solute A in the liquid
(Eq. 10-3)
(Eq. 10-4)
Nx = k^p^c - pA/)
NA = kv (CaJ ~ Cal)
(Eq. 10-5)
(Eq. 10-6)
Nyt — Koc(p*C _ P*)
N* = KOL (C* — Cal)
10-10
-------
k, | k,
Total resistance
Figure 10-4. Resistance to motion encountered by a
molecule being absorbed.
An important fact concerning Equations 10-5 and 10-6 is that they impose an upper limit
on the amount of solute that can be absorbed. The rate of mass transfer depends on the con-
centration departure from equilibrium in either the gas (p^c~ p*) or liquid (c* - c^x) phase.
The larger these concentration differences are, the greater the rate of mass transfer. If
equilibrium is ever reached (p/ic=p* or cal = c*), absorption stops and no net transfer occurs.
Thus, the equilibrium concentrations determine the maximum amount of solute that is
absorbed.
At equilibrium, the overall mass-transfer coefficients are related to the individual mass-
transfer coefficients by the equations below.
(Eq-10_7) + f
(E<"10-8) "ST" T+lk
H' is Henry's law constant (the slope of the equilibrium). Equations 10-7 and 10-8 are useful
in determining which phase controls the rate of absorption. From Equation 10-7, if H' is very
small (which means the gas is very soluble in the liquid), then Koc ~ kj, and absorption is said
to be gas-film controlled. The major resistance to mass transfer is in the gas phase. Con-
versely, if a gas has limited solubility, H' is large, and Equation 10-8 reduces to Kox = k/. The
mass-transfer rate is liquid-film controlled and depends on the solute's dispersion rate in the
liquid phase. Most systems in the air pollution control field are gas-phase controlled since che
liquid is chosen so that che solute will have a high degree of solubility.
10-11
-------
The discussion so far has been based on the two-film theory of absorption. Other theories
offer different descriptions of gas molecule movement from the gas to the liquid phase. Some
of the significant mass-transfer models follow. For these theories, the mass-transfer rate equa-
tion does not differ from that of the two-film model. The difference lies in the way they
predict the mass-transfer coefficient. It has been shown that the rate of mass transfer depends
on a concentration difference multiplied by a resistance factor. Like most theories describing
how something functions, absorption theories provide a basic understanding of the process,
but due to the complexities of "real life" operations, it is difficult to apply them directly. Con-
centrations can easily be determined from operating (c and p) and equilibrium (c* and p*)
data of the system. Mass-transfer coefficients are very difficult to determine from theory.
Theoretically predicted values of the individual mass-transfer coefficients (k, and k,) based on
the two-film theory, do not correlate well with observed values. Overall mass-transfer coeffi-
cients are more easily determined from experimental or operational data. However, the overall
coefficients apply only when the equilibrium line is straight.
Mass-Transfer Models*
Film Theory (Whitman 1923) —First, and probably the simplest, theory proposed for mass
transfer across a fluid. Details of this model are discussed in the text because it is the most
widely used.
Penetration Theory (Higbie 1935) —Assumes that the liquid surface in contact with the gas
consists of small fluid elements. After contact with the gas phase, the fluid elements return to
the bulk of the liquid and are replaced by another element from the bulk-liquid phase. The
time each element spends at the surface is assumed to be the same.
Surface-Renewal Theory (Danckwerts 1951) —Improves on the penetration theory by sug-
gesting that the constant exposure time be replaced by an assumed time distribution.
Film-Penetration Theory (Toor and Marchello 1958) —Combination of the film and penetra-
tion theories. Assumes that a laminar film exists at the fluid interface (as in the film theory),
but further assumes that mass transfer is an unsteady process.
Mass-transfer coefficients are often expressed by the symbols KoCa, k,a, etc., where "a"
represents the surface area available for absorption per unit volume of the column. This
allows for easy determination of the column area required to accomplish the desired separa-
tion. These mass-transfer coefficients are developed from experimental data and are usually
reported in one of two ways: as an empirical relationship based on a function of the liquid
flow, gas flow, or slope of the equilibrium line; or correlated in terms of a dimensionless
number, usually either the Reynolds or Schmidt Number. Figure 10-5 compares the effect on
the mass-transfer coefficient for SOs in water using two types of packing materials (Perry
1973). Packing A consists of one-inch rings and packing B consists of three-inch spiral tiles.
Similar figures are used extensively to compare different absorbers or similar absorbers with
varying operating conditions. It should be noted that these estimated mass-transfer coefficients
are system and packing-type dependent and, therefore, do not have widespread applicability.
The Chemical Engineers' Handbook gives a comprehensive listing of empirically derived coef-
ficients. In addition, manufacturers of packed and plate towers have graphs in their literature
similar to the one in Figure 10-5.
"Source: Diab and Maddox 1982.
10-12
-------
o
*
Where: Packing A= 1-in. rings
Packing B = 3-in. spiral tiles
-L
J.
20
40 80
100 200
G', lb/hr»ftl
400 500 1000
Source: Perry 1973.
Figure 10-5. Comparison of overall absorption coefficient
for SO, in water.
Although the science of absorption is considerably developed, much of the work in practical
design situations is empirical in nature. The following section will apply the principles dis-
cussed to the design of gas absorption equipment. Emphasis has been placed on presenting
information that can be used to estimate absorber size and liquid flow rate.
Review Exercise
1. In the double-resistance, or two-film, theory, a
zone exists thar includes a g^s and liquid
phase on either side of the interface.
a. soluble
b. mass-transfer v
c. droplet
2. True or False? The two-film theory implies that all resis-
tance to movement occurs when the molecule (gaseous
pollutant) is diffusing through the gas and liquid films. \
1. b. mass-transfer
3. In absorption equations, the concentration of a gaseous
pollutant is usually expressed by its
a. diffusion rate. y
b. total pressure.
c. partial pressure.
2. True
3. c. partial pressure.
10-13
-------
4. In calculating the rate of mass transfer of pollutant A, N^,
using the equation = Kog(P-4g— p*)> the term Koc is the
a. equilibrium concentration of pollutant A. x
b. mass-transfer coefficient for the gas film.
c. mass-transfer coefficient for the liquid film.
d. overall mass-transfer coefficient based on the gas phase.
5. True or False? Overall mass-transfer coefficients are only
valid when a plot of the equilibrium data yields an equili-
brium line that is straight. 1,
4. d. overall mass-transfer
coefficient based on
the gas phase.
5. True
Procedures
The effectiveness of an absorption system depends on the solubility of the gaseous contami-
nant. For very soluble gases, almost any type of absorber will give adequate removal.
However, for most gases, only absorbers that provide a high degree of turbulent contact and a
long residence time are capable of achieving high absorption efficiencies. The two most com-
mon high-efficiency absorbers are plate and packed towers. Both of these devices are used
extensively to control gaseous pollutants. Absorber design calculations presented in this lesson
will focus on these two devices.
Numerous procedures are used to design an absorption system. These procedures range in
difficulty and cost from short-cut "rules of thumb" equations to in-depth design procedures
based on pilot plant data. Procedures presented here will be based on the short-cut "rules of
thumb." The approaches discussed in this lesson are for single component systems (i.e., only
one gaseous pollutant).
To design an absorption system, certain parameters are set by either operating conditions or
regulations. The gas stream to be treated is usually the exhaust from a process in the plant.
Therefore, the volume, temperature, and composition of the gas stream are given parameters.
The oudet composition of the contaminant is set by the emission standard which must be met.
The temperature and inlet composition of the absorbing liquid are also usually known. The
main unknowns in designing the absorption system are:
• the flow rate of liquid required,
• the diameter of the vessel needed to accommodate the gas and liquid flow, and
• the height of absorber required to achieve the needed removal.
Procedures for estimating these three unknowns will be discussed in the following sections.
Material Balance
In designing or reviewing the design of an absorption control system, the first task is to deter-
mine the flow rates and composition of each stream entering the system. From the law of con-
servation of mass, the material entering a process must either accumulate or exit. In other
words, "what comes in must go out." A material balance is used to help determine flow rates
and compositions of individual streams. Figure 10-6 illustrates the material balance for a
typical countercurrent-flow absorber. The solute is the "material" in the material balance.
10-14
-------
(Liquid in)
Y
(Gas in)
(Liquid out)
Figure 10-6. Material balance for countercurrent-flow absorber.
The following procedure to set up a material balance and determine the liquid flow rate
will focus on a countercurrent gas-liquid flow pattern. This is the most common flow pattern
used to achieve high-efficiency gas absorption. For cocurrent flow, only a slight modification
of this procedure is required. Equations for crosscurrent flows are very complicated since they
involve a gradient pattern that changes in two directions. They will not be presented here.
X = mole fraction of solute in pure liquid
Y = mole fraction of solute in inert gas
Lm = liquid flow rate, g-mol/h (lb-mole/hr)
= gas flow rate, g-mol/h (lb-mole/hr)
Engineering design work is usually done on a solute-free basis (X, Y) to make the material
balance calculations easier. The solute-free basis is defined in Equations 10-9 and 10-10.
In air pollution control systems, the percent of pollutant transferred, y and x, is generally
small. Therefore, from Equations 10-9 and 10-10, Y = y and X = x. In this lesson, it is
assumed that X and Y are always equal to x and y respectively. If y (inlet gas concentration)
ever becomes larger than a few percent by volume, this assumption is invalid and will cause
errors in the material balance calculations.
(Eq. 10-9)
(Eq. 10-10)
10-15
-------
An overall mass balance across the absorber in Figure 10-7 yields Equation 10-11
(Eq. 10-11) lb m°le in = lb mole out
Gm (in) + U, (in) = Gm (out) + U (out).
Slope of the
operating line =
Driving
forces
X
Figure 10-7. Typical operating line diagram.
For convenience, the top of the absorber is labeled as point 2 and the bottom as point 1. This
changes Equation 10-11 to Equation 10-12.
(Eq. 10-12) Gml + Lnj =
In this same manner, a material balance for the contaminant to be removed is obtained as
expressed in Equation 10-13.
(Eq. 10-13) Gml Y, + X2 = Gm2 Y2 + Lml X,
Equation 10-13 can be simplified by assuming that as the gas and liquid streams flow through
the absorber, their total mass does not change appreciably (i.e., Gml = Gm2 and Lmi = Lm2).
This is justifiable for most air pollution control systems since the mass flow rate of pollutant is
very small compared to the liquid and gas mass flow rates. For example, a 10,000-cfm
exhaust stream containing 1000 ppm S02 would be only 0.1% S02 by volume, or 1.0 cfm. If
10-16
-------
the scrubber were 100% efficient, the gas mass flow rate would change from 10,000 cfm at
Gmi to 9999 cfm at Gm2. The transfer of a quantity this small is negligible in an overall
material balance. Therefore, Equation 10-13 can be reduced to Equation 10-14.
(Eq. 10-14) G.(Y,-Y2) = U(Xl-X2)
By rearranging, Equation 10-14 becomes Equation 10-15.
(Eq. 10-15) Y,-Y2= ~ (X,-X2)
Gm
Equation 10-15 is the equation of a straight line. When this line is plotted on an equilibrium
diagram, it is referred to as an operating line. This line defines operating conditions within
the absorber: what is going in and what is coming out. An equilibrium diagram with a typical
operating line plotted on it is shown in Figure 10-7. The slope of the operating line is the
liquid mass flow rate divided by the gas mass flow rate, which is the liquid-to-gas ratio, or
Lm/Gm. The liquid-to-gas ratio is used extensively when describing or comparing absorption
systems. Determining the liquid-to-gas ratio is discussed in the next section.
Determining the Liquid Requirement
In the design of most absorption columns, the quantity of exhaust gas to be treated (Gm) and
the inlet solute (pollutant) concentration (Y^ are set by process conditions. Minimum accep-
table standards specify the outlet pollutant concentration (Y2). The composition of the liquid
flowing into the absorber (X2) is also generally known or can be assumed to be zero if it is not
recycled. By plotting this data on an equilibrium diagram, the minimum amount of liquid
required to achieve the required outlet pollutant concentration (Y2) can be determined.
Figure 10-8a is a typical equilibrium diagram with operating points plotted for a
countercurrent-flow absorber. At the minimum liquid rate, the inlet gas concentration of
solute (YJ is in equilibrium with the outlet liquid concentration of solute (Xmax). The liquid
leaving the absorber is saturated with solute and can no longer dissolve any more solute unless
additional liquid is added. This condition is represented by point B on the equilibrium curve.
The slope of the line drawn between point A and point B represents the operating condi-
tions at the minimum flow rate in Figure 10-8b. Note how the driving force decreases to zero
at point B. The slope of line AB is (Lm/G„)min, and may be determined graphically or from
the equation for a straight line. By knowing the slope of the minimum operating line, the
minimum liquid rate can easily be determined by substituting in the known gas flow rate.
This procedure is illustrated in Example 10-2.
Determining the minimum liquid flow rate, (Lm/Gm)min, is important since absorber
operation is usually specified as some factor of it. Generally, liquid flow rates are specified at
25 to 100% greater than the required minimum. Typical absorber operation would be 50%
greater than the minimum liquid flow rate (i.e., 1.5 times the minimum liquid-to-gas ratio).
Setting the liquid rate in this way assumes that the gas flow rate set by the process does not
change appreciably. Line AC in Figure 10-8c is drawn at a slope of 1.5 times the minimum
Lm/Gm. Line AC is referred to as the actual operating line since it describes absorber
operating conditions.
10-17
-------
Xt, Y, (I A
CQ
bo
v
3
c
o
u
JJ
o
£
Y,
Driving
forces
(b)
Slope of the
minimum
U
operating line = — (minimum;
Gm
x,
X, mole fraction of solute in liquid phase
(c)
Slope of the
actual l
operating line= — (actual)
G„
Figure 10-8. Graphic determination of liquid flow rate.
10-18
-------
The following example problem illustrates how to compute the minimum liquid rate
required to achieve a desired removal efficiency.
Example 10-2
Using the data and results from Example 10-1, compute the minimum liquid rate of pure
water required to remove 90% of the S02 from a gas stream of 84.9 mVmin (3000 acfm) con-
taining 3% S02 by volume. The temperature is 293 K and the pressure is 101.3 kPa.
Solution
1. First, determine the mole fractions of the pol-
lutants in the gas phase, Y! and Y2, and then
sketch and label the drawing of the system as
shown in Figure 10-9. Yt = 3% S02 by volume
= 0.03
Y2 = 90% reduction of SOz from inlet
concentration
= (10%)(Y1)
= (0.10)(0.03)
= 0.003
Y, = 0.003
L= ?
= 0
Q_= 84.9 mVmin
f Y, = 0.03
Figure 10-9. Material balance for Example 10-2.
2. At the minimum liquid flow rate, the gas mole
fraction going into the absorber, Y1( will be in
equilibrium with the liquid mole fraction leav-
ing the absorber (the liquid will be saturated
with S02). At equilibrium:
Y^H'X, Yi = H'Xi
X = Xi
and Henry's law constant is ,
mole fraction of S02 in air 0.03
H' =42.7
mole fraction of S02 in water 42.7
from Example 10-1. =0.000703
10-19
-------
3. The minimum liquid-to-gas ratio is calculated
by using Equation 10-15.
Y,-Yf = (X,-X2)
U
Gm
0.03-0.003
0.000703-0
Therefore,
Y.-Y,
= 38.4
U
Gm
g-mol of water
g-mol of air
x,-x2
4. Convert the exhaust stream flow rate, Q, to
the exhaust gas molar flow rate, G„ (from
units of m3/min to units of g-mol/min). At
0°C and 101.3 kPa, there are 0.0224 mVg-mol
(for an ideal gas).
At 20 °C:
/293\
0.0224 m3/g-moI ("273 / = 0-024 mVg-mol of ai
Therefore,
air Gm = 89.4 m3/min
1 g-mol
G„ = Q.
1 g-mol of air
0.024 m3
yO.024 m3
= 3538 g-mol of air/min
5. Calculate the minimum liquid flow rate, Lm.
The minimum liquid-to-gas ratio was
calculated in step 3.
Lm _ 4 g-mol of water
Gm g-mol of air
Therefore,
Lm = Gm(38.4)
L_= (3538 g"n°' °f airVs8.4 ^m"' °fwa'C
g-mol of air
mm
=136,000
g-mol of water
min
=136.0
kg-mol of water
mm
Converting to mass units:
kg-mol \/ 18 kg \
min / \ kg-mol/
=136.0
= 2448 kg/min
10-20
-------
6. Figure 10-10 illustrates the graphical solution
for this problem. To obtain the actual operat-
ing line, multiply the slope of the minimum
operating line by 1.5.
Slope of AC = 1.5 slope of AB AC = 1.5(38.4)
= 57.6
0.03
u
Actual
operating line
o
<" 0.02
O
e
o
u
A3
£
| o.oi
>
Minimum operating line
0.0006
0.0008
0.0004
X, mole fraction of SO, in water
0.0002
Figure 10-10. Graphical solution to Example 10-2.
10-21
-------
Review Exercise
1. In absorption calculations, a(an)
equates the
gas and liquid concentrations coming into the absorber
with the gas and liquid concentrations going out of the
absorber.
a. material balance ' /
/ '
b. energy balance c
c. transfer unit
2. In air pollution_ oalculations, the mass of the pollutant is
1. a. material balance
usually very small/large compared to the mass of exhaust
gas being treated and the mass of the liquid used in the
absorber.
In the graph below, the line AB is the
a. equilibrium line.
b. actual operating line.
c. minimum operating line.
2. small
4J
3
S
C C9
O J=
'Z a.
S s
IV}
JU c
o
E
Driving'
X. mole fraction of solute in liquid phase
4. The slope of the actual operating line is
a. minimum liquid-to-gas ratio.
b. Gm/Lm (actual).
c. Lm/Gm (actual).
3. c. minimum operating
line.
4. c. Lm/Gm (actual).
10-22
-------
5. True or False? In the following figure, point B represents
absorber conditions where the liquid leaving the absorber
is saturated with the pollutant and can no longer absorb
any additional pollutant, unless more liquid is added.
3
S
o
E
Y,
JS
a.
to
X,
X, mole fraction of solute in liquid phase
5. True
Sizing a Packed Tower
Packed Tower Diameter
The main parameter affecting the size of a packed column is the gas velocity at which liquid
droplets become entrained in the exiting gas stream. Consider a packed column operating at
set gas and liquid flow rates. By decreasing the diameter of the column, the gas flow rate
(m/s or ft/sec) through the column will increase. If the gas flow rate through the column is
gradually increased (by using smaller and smaller diameter columns), a point will be reached
where the liquid flowing down over the packing begins to be held in the void spaces between
the packing. This gas-to-liquid ratio is termed the loading point. The pressure drop of the
column begins to increase and the degree of mixing between the phases decreases. A further
increase in gas velocity will cause the liquid to completely fill the void spaces in the packing.
The liquid forms a layer over the top of the packing and no more liquid can flow down
through the tower. The pressure drop increases substantially, and mixing between the phases
is minimal. This condition is referred to as flooding, and the gas velocity at which it occurs is
the flooding velocity. Using an extremely large-diameter tower would eliminate this problem.
However, as the diameter increases, the cost of the tower increases.
Normal practice is to size a packed column diameter to operate at a certain percent of the
flooding velocity. A typical operating range for the gas velocity through the columns is 50 to
75% of the flooding velocity. It is assumed that, by operating in this range, the gas velocity
will also be below the loading point.
10-23
-------
A common and relatively simple procedure for estimating flooding velocity (thus, setting a
minimum column diameter) is to use a generalized flooding and pressure drop correlation.
One version of the flooding and pressure drop relationship in a packed tower is in the
Sherwood correlation, shown in Figure 10-11 (Calvert et al. 1972). This correlation is based on
the physical properties of the gas and liquid streams and tower packing characteristics. The
procedure to determine the tower diameter is given below.
¦s-
M
6
bo
••
0/
0.05 —
0.001
— Pressure drop, m HtO/m of packing (in. HjO/ft of packing) —
0.02 —
0.01 —
0.005 —
0.002 —
0.01 0.02 0.05 0.1 0.2
0.5 1
y \ / ,0-5
L\/ G*\
G/Ve
(dimensionless)
Source: Calvert et al. 1972.
Figure 10-11. Generalized flooding and pressure drop correlation.
1. Calculate the value of the abscissa.
(Eq. 10-16) Abscissa
Where: L and G = mass flow rates: any consistent set of units may be used as long as
the term is dimensionless
gg = density of the gas stream
q, = density of the absorbing liquid
2. From the point calculated in Equation 10-16, proceed up the graph to the flooding line
and read the ordinate e.
10-24
-------
3. Rearrange the equation of the ordinate and solve for G'.
(Eq. 10-17) G' =
(e)(e,)(e<)(gc)
0 5
F(^>^0.2
Where: F= packing factor given in Table 10-5 for different types of packing
(Bhatia 1977)
= ratio of specific gravity of the scrubbing liquid to that of water
fh = viscosity of liquid
G' =mass flow rate of gas per unit cross-sectional area of column,
g/s»m2 (lb/sec*ft2)
Qt = density of the absorbing liquid, kg/m3 (lb/ft3)
gt = density of the gas stream, kg/m3 (lb/ft3)
gc = gravitational constant, 9.82 m/s2 (32.2 ft/sec2)
4. G' at operating conditions is a fraction of G' at flooding conditions.
(Eq. 10-18) G 'operating = (f ) (G 'flooding)
Where: f=the percent of flooding velocity, usually 50 to 75%
5. The cross-sectional area of column A is calculated from Equation 10-19.
(Eq. 10-19) A=—-
v' operating
6. The diameter of the column is obtained from Equation 10-20.
(Eq. 10-20) d, =
4A j0-s
= 1.13 A05
10-25
-------
Table 10-5. Packing data.*
Packing
Size
(in.)
Weight
(lb/ft1)
Surface area, a
(ft'/ft* packing
volume)
Void
fraction
(96)
Packing
factor, F
(ft'/ft1)
Price
<$/fts)
Raschig rings
a
52
114
65
580
14.00°, 19.00"
(ceramic
1
44
58
70
155
7.00°. 9.30"
and
lfc
42
36
72
95
6.30", 8.40"
porcelain)
2
38
28
75
65
6.20", 8.30"
3
34
19
77
37
6.50°, 9.50"
Raschig rings
Vt x 1/32
77
128
84
300
105.50
(steel)
1 x 1/32
40
63
92
115
75.10
2x 1/16
38
31
92
57
39.10
Berl saddles
V*
55
274
63
900
—
(ceramic
X
54
155
64
240
24.00°, 32.00"
and
1
48
79
68
110
9.50°, 12.00"
porcelain)
2
38
32
75
45
7.00°, 9.30'
Intalox saddles
y*
54 ¦
300
75
725
—
(ceramic)
a
45
190
78
200
28.35
i
44
78
77
98
10.85
2
42
36
79
40
—
Intalox saddles
1
6.00
63
91
30
—
(plastic)
2
3.75
33
93
20
7.65
3
3.25
27
94
15
—
Pall rings
5/8
7.0
104
87
97
(plastic)
1
5.5
63
90
52
10.75
2
4.5
31
92
25
—
Pall rings
5/8x0.018
38
104
93
73
81.25'
(metal)
thick
1V6 X 0.03
24
39
95
28
33.23'
thick
Tellerettes
1
7.5
55
87
40
12.50
2
3.9
38
93
20
7.30
3
5.0
30
92
15
5.25
'Prices for ceramic packing.
'Prices for porcelain packing.
'Prices for 304 S.S. packing.
'Note: Data for guide purposes only.
Source: Bhatia 1977.
10-26
-------
Example 10-3
This example illustrates the use of Figure 10-11 for computing the minimum allowable
diameter for a packed tower. For the scrubber in Example 10-2, determine the column
diameter if the operating liquid rate is 1.5 times the minimum. The gas velocity should be no
greater than 75% of the flooding velocity, and the packing material is two-inch ceramic
Intalox saddles.
Solution
1. From Example 10-2, the gas molar flow rate in
the absorber, G„, was 3538 g-mol/min and the
minimum liquid flow rate, Lm, was 2448
kg/min. The actual liquid flow rate in the
absorber should be 1.5 times the minimum
flow rate:
L = Lmx 1.5.
Assuming the molecular weight of the exhaust
gas is 29 kg/mol, the gas molar flow rate in
mass units would be:
L=U.X 1.5
= (2448 kg/min) (1.5)
= 3672 kg/min
G = Gm X (29 kg/kg-mol).
G=(3538 g-mol/min)(29 kg/kg-mol)
= (3.538 kg-mol/min)(29 kg/kg-mol)
= 102.6 kg/min
2. Using Equation 10-16, calculate the abscissa
for Figure 10-11.
The densities of air and water at 30 °C are:
= 1.22
qs = 1.17 kg/m3
Qi = 1000 kg/m3.
10-27
-------
3. Using Figure 10-12, with the abscissa of 1.22,
move up to the flooding line and read the
value of e on the ordinate. e = 0.019
a.
¦s-
tl.
¦€
O
bo
di
ol
0.019
0.005
0.002 —
0.001
0.01 0.1 0.5 1.22 2 5 10
Source: Calvert et al. 1972. \G/\ Qt
Figure 10-12. Generalized flooding and pressure drop correlation for Example 10-3.
4. Use Equation 10-17 to calculate the superficial
flooding velocity, G'. The superficial flooding
velocity is the flow rate per unit of cross-
sectional area of the tower.
G' =
(<0(e,)(e«)(gc)
F 0^0-2
For water: 0=1.0
(i, = 0.0008 Pa*s
From Table 10-3, for two-inch Intalox saddles:
F = 131 m2/m3
gc = 9.8 m/'s2
G' =
(0.019)(1.17)(1000) (9.82)
(131)(1)(0.0008)° 2
2.63 kg/m3»s at flooding
0-5
10-28
-------
5. The superficial gas velocity at operating con-
ditions is obtained by using Equation 10-18.
G'
Use 75% of the flooding velocity for f.
G'op.~t.n, = (0.75)(2.63)
= 1.97 kg/m2*s
6. Calculate the cross-sectional area of the packed
tower by using Equation 10-19.
G
A =
G'
A _ (102.6 kg/min)(l min/60 sec)
A — 111 1
1.97 kg/m2»s
= 0.87 m2
7. Calculate the tower diameter by using Equa-
tion 10-20.
d, = I
d* =
4(0.87)
3.14
0-5
Where:
7T= 3.14
= 1.05 m
*1.1 m
8. Figure 10-11 may also be used to estimate the
pressure drop across the absorber, Ap, once
the superficial gas velocity for operating con-
ditions has been set. First, plug G'n,
back
into Equation 10*17 and rearrange the equa-
tion to get the ordinate, e.
e =
G'*4>Fji,
Q*Q<&
e =
(1.97)* (1) (131) (0.0008)02
The abscissa is equal to 1.22.
Then, from Figure 10-11, read Ap.
(1.17)(1000)(9.82)
= 0.0106
ordinate = 0.0106
abscissa = 1.22
Ap = 0.416 m of water/m of packing
10-29
-------
Packed Tower Height
The height of a packed column refers to the depth of packing material needed to accomplish
the required removal efficiency. The more difficult the separation, the larger the packing
height required. For example, a much larger packing height would be required to remove S02
than to remove CI from an exhaust stream using water as the absorbent. This is because CI is
more soluble in water than SO?. Determining the proper height of packing is important since
it affects both the rate and efficiency of absorption.
A number of theoretical equations are used to predict the required packing height. These
equations are based on diffusion principles. Depending on which phase is controlling the
absorption process, either Equation 10-5 or 10-6 is used as the starting point to derive an
equation to predict column height. A material balance is then set up over a small differential
section of the column.
The general form of the design equation for a gas-phase controlled resistance is given in
Equation 10-21.
Y,
G' [ dY
(Eq. 10-21) Z- RocaP ] (!_Y)(Y-Y*)
Yj
Where: Z = height of packing, m
G' = mass flow rate of gas per unit cross-sectional area of column, g/s»m2
Koc= overall mass-transfer coefficient based on the gas phase, g-mol/h*m2*Pa
a = interfacial contact area, m1
P = pressure of the system, kPa
In analyzing Equation 10-21, the term G'/KoCaP has the dimension of meters and is
defined as the height of a transfer unit. The term inside the integral is dimensionless and
represents the number of transfer units needed to make up the total packing height. Using the
concept of transfer units, Equation 10-21 can be simplified to:
(Eq. 10-22) Z = HTU x NTU
Where: HTU = height of a transfer unit, m
NTU = number of transfer units.
The concept of a transfer unit comes from the assumptions used in deriving Equation
10-21. These assumptions are: (1) that the absorption process is carried out in a series of con-
tacts, or stages, and (2) that the streams leaving these stages are in equilibrium with each
other. These stages can be visualized as individual transfer units. The total tower height is
equal to the number of transfer units times the height of each unit. Although a packed col-
umn operates as one continuous separation (differential contactor) process, in design termi-
nology it is treated as discrete sections (transfer units). The number and the height of a
transfer unit are based on either the gas or the liquid phase. Equation 10-22 now becomes:
(Eq. 10-23) Z = NocHog = N0i Ho£
Where: Noc = number of transfer units based on an overall gas-film coefficient
N0i = number of transfer units based on an overall liquid-film coefficient
Hoc = height of a transfer unit based on an overall gas-film coefficient, m
H0£ = height of a transfer unit based on an overall liquid-film coefficient, m.
10-30
-------
The number of transfer units, NTU, can be obtained experimentally or calculated from a
variety of methods. For the case where the solute concentration is very low and the equili-
brium line is straight, Equation 10-24 can be used to determine the number of transfer units
(Noc) based on the gas-phase resistance. Equation 10-24 can be derived from the integral por
tion of Equation 10-21.
Gm = molar flow rate of gas, kg-mol/h
Lm = molar flow rate of liquid, kg-mol/h
X* = mole fraction of solute entering the column
Y, = mole fraction of solute in entering gas
Yj = mole fraction of solute in exiting gas
Equation 10-24 may be solved directly or graphically by using the Colbum diagram, which is
presented in Figure 10-13. The Colbum diagram is a plot of the Noc versus ln[Y, — mX2/Y2 - mX2]
at various values of (mGm/L„). The term (mGm/Lm) is referred to as the absorption factor.
Figure 10-13 is used by first computing the value of [Yi — mX?/Yj — mX2], reading up the graph
to the line corresponding to (mGw/U,), and then reading across to obtain the Noc.
Equation 10-24 can be further simplified for situations where a chemical reaction occurs or
if the solute is extremely soluble. In these cases, the solute exhibits almost no partial pressure;
therefore, the slope of the equilibrium line approaches zero (m—0). For either of these cases,
Equation 10-24 reduces to Equation 10-25.
The number of transfer units depends only on the inlet and outlet concentration of the solute.
For example, if the conditions of Equation 10-25 are met, achieving 90% removal of any
pollutant requires 2.3 transfer units. Equation 10-25 applies only when the equilibrium line is
straight and the slope approaches zero (for very soluble or reactive gases).
(Eq. 10-24)
Where: m = slope of equilibrium line
(Eq. 10-25)
2
10-31
-------
Source: Perry 1973.
Yi — mXt
Y, - mX,
Figure 10-13. Colbuni diagram.
Values for the height of a transfer unit used in designing absorption systems are usually
obtained from experimental data. To ensure greatest accuracy, vendors of absorption equip-
ment normally perform pilot plant studies to determine the Hi U. For common absorption
systems, such as NHS and water, manufacturers have developed graphs to use for estimating
HTU. These graphs do not provide the accuracy of pilot plant data, but they are less expen-
sive and easier to use. Figure 10-14 gives a typical example of these graphs for an ammonia
and water system. In this figure, the superficial liquid flow rate is plotted versus the Hoc
with the superficial gas rate as a parameter. It is also common to plot gas rate versus the
Hoc and have the liquid rate as a parameter. Additional information on other gas-liquid
systems can be found in Chemical Engineers' Handbook (Perry 1973). In applying these
data, process conditions must be similar to conditions at which the HTU was measured.
10-32
-------
3.6
a
3?
2.0
2.8
1.2
G' = 500 lb/hr»ft2
Where: O = lV£-in. Raschig rings
A= 1-in. Tellerettes
0.4
500
1000
L', lb/hr«ftl
1500
2000
Figure 10-14. Column packing comparison for ammonia and water system.
When no experimental data are available, or if only a preliminary estimate of absorber effi-
cieny is needed, generalized correlations are available to predict the height of a transfer unit.
The correlations for predicting the Hoc or the Hot are empirical in nature and are a function
1. type of packing,
2. liquid and gas flow rates,
3. concentration and solubility of the pollutant,
4. liquid properties, and
5. system temperature.
These correlations can be found in engineering texts such as Chemical Engineers' Handbook
(Perry 1973), Wet Scrubber System Study, Volume I (Calvert et al. 1972), or Mass Transfer
Operations (Treybal 1968). For most applications, the height of a transfer unit ranges between
0.3 and 1.2 m (1 to 4 ft) (Calvert 1977). As a rough estimate, 0.6 m (2.0 ft) can be used.
of:
10-33
-------
Example 10-4
From pilot plant studies of the absorption system in Example 10-2 it was determined that the
Hog for the SOr water system is 0.829 m (2.72 ft). Calculate the total height of packing
required to achieve 90% removal. The following data were taken from the previous examples.
m
= 42.7 kg~mo* water Henry's law constant for the equilibrium diagram for S02
kg-mol of air an(j water System (Example 10-1)
G„, = 3.5 kg-mol/min
kg kg-mol
L*. = 3672 r— x -f—
mm lo kg
= 204 kg-mol/min
X2 = 0 (no recycle liquid)
Y, = 0.03
Y? = 0.003
Solution
1. Calculate the number of transfer units, Nog,
by using Equation 10-24.
In
— mXt\f _ mG„\ + mG.
Y, — mXJ\ L*t / Lm
In
Nog — — — ¦ Nog =
0.03 \ ^ _ (42.7)(3.5) \ + (42.7)(3
0.003/ V 204 / 204
l_ mGm " (42.7)(3.5)
U 204
= 5.04
2. The total packing height can be calculated
using Equation 10-23.
Z = Hoc x Nog Z = (0.829)(5.04)
= 4.18 m of packing height
10-34
-------
Review Exercise
1. In designing a packed tower, the normal practice is to
make the tower diameter so that the unit will operate at
of rhe flooding velnriry rare.
a. 50 to 75%
b. 100%
c. 150%
2. True or False? The Sherwood correlation can be used to
calculate the tower diameter of a packed tower, if the
minimum liquid rate, L*, and the gas flow rate, G,
through the absorber are known.
1. a. 50 to 75%
3. In estimating packing height in a packed tower, the pack-
ing sections are broken up into discrete sections called
a. transfer units.
b. gas-film coefficients. r ^
c. liquid-film coefficients.
2. True
4. The packing height, Z, can be estimated from
Z = HTU x NTU
What are the terms HTU and NTU?
3. a. transfer units.
5. True or False? The Colbum diagram can be used to
estimate the number of transfer units based on an overall
gas-film coefficient, Nog, if the absorption factor,
mGm/Lm, the inlet and outlet pollutant concentrations,
and the liquid recycle concentrations are known. '
4. HTU = height of a
transfer unit
NTU = number of
transfer units
6. The height of a transfer unit is a function of
a. type of packing.
b. liquid and gas flow rates.
c. pollutant concentration and solubility.
d. liquid properties and system temperature. ^
e. all of the above
5. True
(by using Figure 10-13)
7. For most packed tower applications, the height of a trans-
fer unit can be estimated to be
a. 3 to 4.6 m (10 to 15 ft). X
b. 0.3 to 1.2 m (1 to 4 ft). L
c. 1.82 to 3 m (6 to 10 ft).
6. e. all of the above
7. b. 0.3 to 1.2 m
(1 to 4 ft).
10-35
-------
Sizing a Plate Tower
Another scrubber used extensively for gas absorption is a plate tower. Here, absorption occurs
on each plate, or stage. These are commonly referred to as discrete stages, or steps. The
following discussion presents a simplified method for sizing or reviewing the design plans of a
plate tower. The method for determining the liquid flow rate in the plate tower is the same as
previously discussed. Methods for estimating the diameter of a plate tower and the theoretical
number of plates follow.
Plate Tower Diameter
The minimum diameter of a single-pass plate tower is determined by using the gas velocity
through the tower. If the gas velocity is too fast, liquid droplets are entrained, causing a con-
dition known as priming. Priming occurs when the gas velocity through the tower is so fast
that it causes liquid on one tray to foam and then rise to the tray above. Priming reduces
absorber efficiency by inhibiting gas and liquid contact. For the purpose of determining tower
diameter, priming in a plate tower is analogous to the flooding point in a packed tower. It
determines the minimum acceptable diameter. The actual diameter should be larger.
The smallest allowable diameter for a plate tower is expressed in Equation 10-26.
(Eq. 10-26) d, = iA(Q.Ve7)0-5
Where: Q_= volumetric gas flow, m3/h
ip = empirical correlation, m°"h0S/kg0'"
Qt = gas density, kg/m3
The term is an empirical correlation and is a function of both the tray spacing and the den-
sities of the gas and liquid streams. Values for in Table 10-6 are for a tray spacing of 61 cm
(24 in.) and a liquid specific gravity of 1.05 (Calvert et al. 1972). If the specific gravity of a
liquid varies significantly from 1.05, the values for \p in Table 10-6 cannot be used.
Table 10-6. Empirical constants for Equation 10-26.
Tray
Metric
English \l/"
Bubble cap
Sieve
Valve
0.0162
0.0140
0.0125
0.1386
0.1198
0.1069
"Metric it is expressed in m0l5h0Vkg°'", for
use with Q expressed in mVh, and q,
expressed in kg/m3.
'English ^ is expressed in ft°"min<,Vlb°">
for use with Q,in cfm, and q, expressed in
lb/ft3.
Source: Calvert et al. 1972.
10-36
-------
Depending on operating conditions, trays are spaced with a minimum distance between
plates to allow the gas and liquid phases to separate before reaching the plate above. Trays
should be spaced to allow for easy maintenance and cleaning. Trays are normally spaced 45
to 70 cm (18 to 28 in.) apart. In using Table 10-6 for a tray spacing different from 61 cm, a
correction factor must be used. Figure 10-15 is used to determine the correction factor, which
is multiplied by the estimated diameter. Example 10-5 illustrates how to estimate the
minimum diameter of a plate tower.
c
o
o
U
Source: Calvert et al. 1972.
0.4 0.5 0.6
Tray spacing, m
0.7 0.8
Figure 10-15. Tray spacing correction factor.
Example 10-5
For the conditions described in Example 10-2, determine the minimum acceptable diameter if
the scrubber is a bubble-cap tray tower. The trays are spaced 0.53 m (21 in.) apart.
Solution
1. From Example 10-2 the following information
is obtained:
Gas flow rate, Qr=84.9 m3/min
Gas density, qx= 1.17 kg/m3.
Convert the gas flow rate, Q, to units of m3/h
Q_=(84.9 m3/min)(60 min/hr)
= 5094 m3/h
10-37
-------
2. From Table 10-6, the empirical constant
= 0.0162 m° 2sh° Vkg° The minimum
diameter, d,, of the plate tower can be
estimated by using Equation 10-26.
d, = HQ.^0l)°-5 d, = (0.0162)[5094(vTT7 )]0'5
= 1.2 m
3. The tray spacing for each tray is 0.53 m. Since
Table 10-6 values are for a tray of 0.61 m,
correct the diameter using Figure 10-16.
Read a correction factor of 1.05.
.1
05
0
0.4 0.53 0.6
Tray spacing, m
Figure 10-16. Tray spacing correction factor
for Example 10-5.
4. Adjust the minimum plate tower diameter
value by using the correction factor.
Adjusted d, = d, (from step 2) x correction factor d, = 1.2 m (1.05)
= 1.26 m
Note: The value of 1.26 m is the minimum estimated
tower diameter based on priming conditions. In
practice, a larger diameter based on economic
conditions is usually chosen.
10-38
-------
Number of Theoretical Plates
Several methods are used to determine the number of ideal plates, or trays, required for a
given removal efficiency. These methods, however, can become quite complicated. One
method used is a graphical technique. The number of ideal plates is obtained by drawing
"steps" on an operating diagram. This procedure is illustrated in Figure 10-17. This method
can be rather time consuming, and inaccuracies can result at both ends of the graph.
Equation 10-27 is a simplified method used to estimate the number of plates. This equation
can only be used if both the equilibrium and operating lines for the system are straight. This
is a valid assumption for most air pollution control systems. This equation, taken from
Sherwood and Pigford (1952), is derived in the same manner as Equation 10-24 for computing
the Nog of a packed tower. The difference is that Equation 10-27 is based on a stepwise solu-
tion instead of a continuous contactor, as is the packed tower. (Note: This derivation is
referred to as the height equivalent to a theoretical plate, or HETP instead of HTU.)
This equation is used to predict the number of theoretical plates required to achieve a given
removal efficiency. The operating conditions for a theoretical plate assume that the gas and
liquid streams leaving the plate are in equilibrium with each other. This ideal condition is
never achieved in practice. A larger number of actual trays are required to compensate for
this decreased tray efficiency.
Y
Operating
line
Equilibrium
line
Note: Lines AB-BC arc one theoretical plate.
Need a total of 2.3 plates.
X
Figure 10-17. Graphic determination of the number
of theoretical plates.
(Eq. 10-27)
10-39
-------
Three types of efficiency are used to describe absorption efficiency for a plate tower:
1. an overall efficiency, which is concerned with the entire column,
2. Murphree efficiency, which is applicable with a single plate, and
3. local efficiency, which pertains to a specific location on a plate.
A number of methods are available to predict these plate efficiencies. These methods are com-
plex, and values predicted by two different methods for a given system can vary by as much as
80% (Zenz 1972).
The simplest of tray efficiency concepts, the overall efficiency, is the ratio of the number of
theoretical plates to the number of actual plates. Since overall tray efficiency is an over-
simplification of the process, reliable values are difficult to obtain. For a rough estimate,
overall tray efficiencies for absorbers operating with low-viscosity liquid normally fall in a 65
to 80% range (Zenz 1972).
Example 10-6
Calculate the number of theoretical plates required for the scrubber in Example 10-5 using
the same conditions as those in Example 10-4. Estimate the total height of the column if the
trays are spaced at 0.53-m intervals, and assume an overall tray efficiency of 70%.
Solution
1. From Example 10-5 and the previous
examples, the following data are obtained:
m = 42.7
Yi= 0.03
Ys = 0.003
Xt = 0.0
L* = 204 kg-mol/min
Gm = 3.5 kg-mol/min.
The number of theoretical plates is estimated
by using Equation 10-26.
= 3.94 theoretical plates
10-40
-------
2. Assuming that the overall efficiency of each
plate is 70%, estimate the actual number of
plates.
Actual plates =
estimated plates
70%
Actual plates =
3.94
0.70
= 5.6 or 6 plates
(since you can't have a
fraction of a plate)
3. Estimate the height of the tower by using
Z = Np x tray spacing + top height of tower.
The top height of the tower is the distance
that allows the gas-vapor mixture to separate.
This distance is usually the same distance as
the tray spacing.
Note: This height is approximately the same as that
predicted for the packed tower in Example 10-4.
This seems logical since both packed and plate
towers are efficient gas-absorption devices.
However, due to the many assumptions, no
concrete generalization can be made.
Z = 6 plates (0.53 m) + 0.53 m
= 3.18 + 0.53
= 3.71 m
Review Exercise
1. In a plate tower, if the gas velocity through the tower is
too fast, liquid droplets become entrained in the gas
stream, causing a condition called
a. pumping.
b. streaking.
c. priming.
2. True or False? For the purpose of determining a plate-
tower diameter, priming in a plate tower is the same as
the flooding point in a packed tower.
1. c. priming.
2. True
10-41
-------
3. In a plate tower, the following equation
In
N,=
Y,-mX,W mG„ \ + mG„
Y, — mXi
In
mG,
is used to calculate the
a. number of transfer units based on an overall gas-film
coefficient.
b. number of transfer units based on Henry's law constant.
c. number of theoretical plates.
CI.
4. In plate towers, the efficiency of each plate, or tray, is
usually
a. 20 to 30%.
b. 65 to 80%.
c. 90 to 100%.
3. c. number of theoretical
plates.
4. b. 65 to 80%.
Review of Design Criteria for Permits
The principal design criteria are the exhaust flow to the absorber, measured in units of
m3/min (ft3/min, or acfm), and the gaseous pollutant concentration, measured in units of
ppm. The exhaust volume and pollutant concentration are set by the process exhaust
conditions. Once these criteria are known, the vendor can begin to design the absorber for the
specific application. A thorough review of the design plans should consider the factors
presented below.
Exhaust gas characteristics —average and maximum exhaust flow rates to the absorber,
and chemical properties such as dew point, corrosiveness, pH, and solubility of the pollu-
tant to be removed should be measured or accurately estimated.
Liquid flow—the type of scrubbing liquid and the rate at which the liquid is supplied to
the absorber. If the scrubbing liquid is to be recirculated, the pH and amount of suspended
solids (if any) should be monitored to ensure continuous reliability of the absorbing system.
Pressure drop—the pressure drop (gas-side) at which the absorber will operate; the
absorber design should also include a means for monitoring the pressure drop across the
system, usually by manometers.
PH — the pH at which the absorber will operate; the pH of the absorber should be
monitored so that the acidity or alkalinity of the absorbing liquor can be properly adjusted.
Removal of entrained liquid —mists and liquid droplets that become entrained in the
"scrubbed" exhaust stream should be removed before exiting the stack. Some type of
entrainment separator, or mist eliminator, should be included in the design.
10-42
-------
Emission requirements — collection efficiency in terms of parts per million (ppm) to meet
the air pollution regulations; collection efficiency can be high (90 to 99%) if the absorber is
properly designed. The agency review engineer can use the equations listed in this lesson to
estimate the absorber removal efficiency, liquid flow rate, tower diameter, and packing
height. However, these equations can only estimate these values, and they should not be
used as the basis to either accept or reject the design plans submitted for the permit
process.
Summary
For gas absorption, the two devices most often used are the packed tower and the plate tower.
Both of these devices, if designed and operated properly, can achieve high collection efficien-
cies for a wide variety of gases. Other scrubbing systems can be used for absorption, but are
limited to cases where the gases are highly soluble. For example, spray towers, Venturis, and
cyclonic scrubbers are designed assuming the performance is equivalent to one single
equilibrium stage (i.e., N0c= 1) (Perry 1973).
The equations and procedures used in designing packed and plate towers are very similar.
Both are based on solubility, the mass-transfer model, and the geometry of the tower. The
main difference is that the equations for a plate tower are based on a stepwise process,
whereas those for a packed tower are based on a continuous-contacting process. Care must be
taken when applying any of the equations presented in this lesson (or in other texts). Some of
the equations are empirical and are applicable only under a similar set of conditions. Used
correctly, these procedures can be a useful tool in checking absorber designs or in determining
the effect of a process change on absorber operation.
When checking the design plans for the permit process, the agency engineer should check
its files or another agency's files for similar applications for absorber installations. A review of
these data will help determine if the absorber design specifications submitted by the industrial
source's officials are adequate to achieve pollutant removal efficiency for compliance with the
regulations. The agency engineer should require the source owner/operator to conduct stack
tests (once the source is operating) to determine if the source is in compliance with local,
State, and Federal regulations. The agency engineer should also require that the source
owner/operator submit an operation and maintenance schedule that will help keep the scrub-
ber system on line.
10-43
-------
References
Bhatia, M. V. 1977. Packed tower and absorption design. In Air pollution control and design
handbook. P. N. Cheremisinoff and R. A. Young, eds. New York: Marcel Dekker, Inc.
Bethea, R. M. 1978. Air pollution control technology. New York: Van Nostrand Reinhold Co.
Calvert, S. 1977. How to choose a particulate scrubber. Chem. Eng. 84:54-68.
Calvert, S.; Goldshmid, J.; Leith, D.; and Mehta, D. 1972. Wet scrubber system study,
volume I: scrubber handbook. EPA-R2-72-118c. U.S. Environmental Protection Agency.
Danckwerts, P. V. 1951. Ind. Eng. Chem. 43:1460.
Diab, Y. S., and Maddox, R. N. 1982. Absorption. Chem. Eng. 89:38-56.
Higbie, R. 1935. Transactions of AIChE 31:365.
Joseph, G. T., and Beachler, D. S. December 1981. Control of gaseous emissions. APTI
Course 415, EPA 450/2-81-005. U.S. Environmental Protection Agency.
Marchello, J. M. 1976. Control of air pollution sources. New York: Marcel Dekker, Inc.
McCabe, W. L., and Smith, C. J. 1967. Unit operations of chemical engineering. New York:
McGraw-Hill Book Co.
McDonald, J. W. 1977. Packed wet scrubbers. In Air pollution control and design handbook.
P. N. Cheremisinoff and R. A. Young, eds. New York: Marcel Dekker, Inc.
Perry, J. H., ed. 1973. Chemical engineers' handbook, 5th ed. New York: McGraw-Hill Book
Co.
Sherwood, T. K., and Pigford, R. L. 1952. Absorption and extraction. New York: McGraw-
Hill Book Co.
Theodore, L., and Buonicore, A. J. 1975. Industrial control equipment for gaseous
pollutants, vol. I. Cleveland: CRC Press.
Toor, H. L., and Marchello, J. M. 1958.fournal of AIChE 4:97.
Treybal, R. E. 1968. Mass transfer operations, 2nd ed. New York: McGraw-Hill Book Co.
Whitman, W. G. 1923. Chem. and Met. Eng. 29:147.
Zenz, F. A. 1972. Designing gas absorption towers. Chem. Eng. 79:120-138.
10-44
-------
TECHNICAL REPORT DATA
/Please read Instructions on the reverse before completing)
1 . REPORT NO. 2.
EPA 450/2-82-020
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
APTI Course SI.-412C
Wet Scrubber Plan Review
Self-instructional Guidebook.
5. REPORT DATE
March 1984
6. PERFORMING ORGANIZATION CODE
7. AUTHORIS)
Gerald T. Joseph, P.E., and David S. Beachler
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Northrop Environmental Training
Northrop Services, Inc.
P.O. Box 12313
Research Triangle Park, NC 27709
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-02-3573
12. SPONSORING AGENCY NAME AND AOORESS
U.S. Environmental Protection Agency-
Manpower and Technical Information Branch
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOO COVERED
Self-instructional course
14. SPONSORING AGENCY CODE
13. SUPPLEMENTARY NOTES '
EPA Project Officer for this self-instructional course is R. E. Townsend,
EPA-ERC, MD 20, Research Triangle Park, NC 27711
16. ABSTRACT
This Self-instructional Guidebook is a self-instructional course, APTI Course
SI:412C, Wet Scrubber Plan Review." This course is designed for engineers and
other technical persons responsible for reviewing plans for the installation of
wet scrubbers used to remove particulate and/or gaseous pollutants from
industrial sources. Major topics include: general description of wet scrubbers,
particle collection and absorption theory, estimating collection efficiency,
wet scrubber components, wet scrubbers used in flue gas desulfurization (FGD),
and operation and maintenance problems associated with wet scrubbing systems.
i
17. KEY WOROS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATt Field/Group
Air Pollution Control Equipment
Wet Scrubbers
Particulate Emission Control
Gaseous Emission Control
Self-training Manual
Self-instructional
Guidebook for Wet
Scrubber Plan Review
is. distribution statement Unlimited
National Audio—Visual Center
National Archives and Records Service
GSA Order Service HH
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
205
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
epa Form 2220-1 (s-73) Washington, DC 20409
10-45
------- |