EPA-600/2-77-234
November 1977
Environmental Protection Technology Series
ENERGY UTILIZATION
BY WET SCRUBBERS
Industrial Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental Protection
Agency, have been grouped into five series. These five broad categories were established to
facilitate further development and application of environmental technology. Elimination of
traditional grouping was consciously planned to foster technology transfer and a maximum
interface in related fields. The five series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ENVIRONMENTAL PROTECTION TECHNOLOGY
series. This series describes research performed to develop and demonstrate instrumenta-
tion, equipment, and methodology to repair or prevent environmental degradation from point
and non-point sources of pollution. This work provides the new or improved technology
required for the control and treatment of pollution sources to meet environmental quality
standards.
EPA REVIEW NOTICE
This report has been reviewed by the U.S. Environmental Protection Agency, and approved
for publication. Approval does not signify that the contents necessarily reflect the views and
policy of the Agency, nor does mention of trade names or commercial products constitute
endorsement or recommendation for use.
This document is available to the public through the National Technical Information Service,
Springfield, Virginia 22*61.
-------
EPA-600/2-77-234
November 1977
ENERGY UTILIZATION
BY WET SCRUBBERS
by
Konrad T. Semrau, Clyde L Witham, and William W. Kerlin
SRI, International
333 Ravenswood Avenue
Menlo Park, California 94025
Contract No. 68-02-2103
ROAP 21ADJ-005
Program Element No. 1AB012
EPA Project Officer: Dale L. Harmon
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, N.C. 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, D.C. 20460
-------
ABSTRACT
An experimental pilot-plant investigation was made of the compara-
tive performance of particulate scrubbers that draw contacting power for
gas/liquid contacting from the gas stream and from the liquid stream.
Three synthetic polydisperse test aerosols were used; these had similar
size distributions but different mass-median particle diameters of 1.05,
0.68, and 0.42 |j,m, respectively. The different contactors were compared
on the basis of collection efficiencies obtained at given levels of con-
tacting power, using the performance of an orifice contactor as a
reference level.
A series contactor consisting of staggered, multiple orifices in
series gave essentially the same performance as the single-orifice con-
tactor at the higher contacting power levels but tended to give poorer
performance at the lower power levels. This behavior was accentuated as
the aerosol particle size decreased.
Five different pressure spray nozzles were tested, as was a sixth
spray nozzle in combination with a single orifice. All these contactor
configurations gave poorer performance than did the reference orifice
scrubber at the same total contacting power. However, the deviation of
the efficiency from the reference level proved to be a function of the
fraction of the total contacting power that was derived from the liquid
stream. The relative deviation of the spray scrubber efficiency from
the reference level increased as the aerosol particle size decreased.
All the spray contactors but one gave very similar performances at
comparable conditions despite radical differences in spray nozzle de-
signs and spray configurations.
This report was submitted in fulfillment of Contract No. 68-02-2103
by Stanford Research Institute under the sponsorship of the U.S. Environ-
mental Protection Agency. This report covers the period June 20, 1975,
to December 20, 1976, and work was completed as of December 20, 1976.
iii
-------
CONTENTS
ABSTRACT iii
FIGURES. , vi
TABLES ix
ABBREVIATIONS AND NOMENCLATURE x
ACKNOWLEDGMENT xii
1. INTRODUCTION 1
General 1
Objectives 2
2. SUMMARY AND CONCLUSIONS 3
3. RECOMMENDATIONS 8
4. BACKGROUND 9
5. EXPERIMENTAL METHODS AND EQUIPMENT 12
Pilot-Plant Scrubber 12
Gas/Liquid Contactors 14
Aerosol Generation 14
Aerosol Sampling and Characterization 25
Aerosol Analysis 26
General Scrubber Test Procedure 28
6. RESULTS AND DISCUSSION 29
Aerosol Characterization with the Miniscrubber 29
Test Aerosol Generation 30
Orifice Scrubber 42
Multiple-Orifice Series Scrubber 54
Spray and Spray-Orifice Scrubbers 54
General Discussion 91
Comparison with Previous Studies 91
Performance Curves 92
Particle Size/Efficiency Relationships 93
Series Scrubbing 95
Spray Scrubbing 95
REFERENCES . , 99
iv
-------
APPENDICES
A LITERATURE REVIEW 101
B AEROSOL GENERATOR 106
GLOSSARY 109
CONVERSION FACTORS Ill
-------
FIGURES
Number Page
1 Flowsheet of Experimental Scrubber System 13
2 Orifice Gas/Liquid Contactor 15
3 Multiple-Orifice Series Gas/Liquid Contactor. ... 16
4 Spray Gas/Liquid Contactor--Configuration SSI ... 18
5 Spray Gas/Liquid Contactor--Configuration SS2 ... 19
6 Spray Gas/Liquid Contactor--Configuration SS3 ... 20
7 Spray Gas/Liquid Contactor--Configuration SS4 ... 21
8 Spray Gas/Liquid Contactor--Configuration SS5 ... 22
9 Spray-Orifice Gas/Liquid Contactor 23
10 Electron Micrograph of Aerosol D 33
11 Electron Micrograph of Aerosol E 34
12 Electron Micrograph of Aerosol F 35
13 Electron Micrograph of Aerosol G 37
14 Particle-Size Distribution of Aerosol D--Mass
Basis (by Cascade Impactor) 38
15 Particle-Size Distribution of Aerosol E--Mass
Basis (by Cascade Impactor) 39
16 Particle-Size Distribution of Aerosol G--Mass
Basis (by Cascade Impactor) 40
17 Summary Curves--Particle-Size Distributions of
Aerosols D, E, and G (Mass Basis) 41
18 Variations in Aerosol D Indicated by
Miniscrubber Signature Curves 43
19 Miniscrubber Signature Curves for Aerosols D, E,
F, and G 44
20 Orifice Scrubber Performance Curve for Aerosol D. . 46
21 Orifice Scrubber Performance Curve for Aerosol E. . 48
22 Orifice Scrubber Performance Curve for Aerosol F. . 50
23 Orifice Scrubber Performance Curve for Aerosol G. . 52
24 Summary of Orifice Scrubber Performance Curves
for Aerosols D, E, F, and G 55
vi
-------
Number Page
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
Multiple-Orifice Series Scrubber Performance on
Aerosol D
Multiple-Orifice Series Scrubber Performance on
Aerosol E
Multiple-Orifice Series Scrubber Performance on
Aerosol G
Configuration SSI Spray Scrubber Performance on
Aerosol D
Configuration SS2 Spray Scrubber Performance on
Aerosol D
Configuration SS3 Spray Scrubber Performance on
Aerosol D
Configuration SS4 Spray Scrubber Performance on
Aerosol D
Configuration SS4 Spray Scrubber Performance on
Aerosol G
Configuration SS5 Spray Scrubber Performance on
Aerosol D
Configuration SS5 Spray Scrubber Performance on
Aerosol G
Spray-Orifice Scrubber Performance on Aerosol D . .
Spray-Orifice Scrubber Performance on Aerosol G . .
Performance of Spray Scrubbers on Aerosol D at
Values of f = 0-0.39
Performance of Spray Scrubbers on Aerosol D at
Values of f = 0.40-0.59
Performance of Spray Scrubbers on Aerosol D at
Values of f = 0.60-0.69
Performance of Spray Scrubbers on Aerosol D at
Values of f = 0.70-0.79
Performance of Spray Scrubbers on Aerosol D at
Values of f = 0.80-0.89
Performance of Spray Scrubbers on Aerosol D at
Values of f = 0.90-0.99
Performance of Spray Scrubbers in Ejector
Configuration on Aerosol D
Performance of Spray Scrubbers on Aerosol G at
Values of f = 0.90-0.99
59
60
61
70
71
72
73
74
75
76
77
78
82
83
84
85
86
87
88
89
vii
-------
Number Page
45 Performance of Spray Scrubbers in Ejector
Configuration on Aerosol G 90
46 Performance of Orifice Scrubber as a Function of
Aerosol Particle Size 94
viii
-------
TABLES
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Gas/Liquid Contactors
Pressure Spray Nozzles Used for Gas/Liquid
Contacting
Test Aerosol Characteristics
Performance of Orifice Scrubber on Aerosol D. . . .
Performance of Orifice Scrubber on Aerosol E. . . .
Performance of Orifice Scrubber on Aerosol F. . . .
Performance of Orifice Scrubber on Aerosol G. . . .
Performance of Multiple-Orifice Series Scrubber
on Aerosol D
Performance of Multiple-Orifice Series Scrubber
on Aerosol E
Performance of Multiple-Orifice Series Scrubber
on Aerosol G
Performance of Configuration SSI Spray Scrubber
on Aerosol D
Performance of Configuration SS2 Spray Scrubber
on Aerosol D
Performance of Configuration SS3 Spray Scrubber
on Aerosol D
Performance of Configuration SS4 Spray Scrubber
on Aerosol D
Performance of Configuration SS4 Spray Scrubber
on Aerosol G
Performance of Configuration SS5 Spray Scrubber
on Aerosol D
Performance of Configuration SS5 Spray Scrubber
on Aerosol G
Performance of Spray-^Orifice Scrubber on
Aerosol D
Performance of Spray-Orifice Scrubber on
Aerosol G
Page
5
17
32
45
49
51
53
56
57
58
62
62
63
64
65
66
67
68
69
ix
-------
ABBREVIATIONS AND NOMENCLATURE
ABBREVIATIONS
atm -- atmospheres (pressure)
cm -- centimeters
cm WC -- centimeters of water (pressure)
ft -- feet
g -- grams
gal -- gallons
hp -- horsepower
kWh -- kilowatt hours
m -- meters
rag -- milligrams
min -- minutes
mm -- millimeters
N -- normality (of solution)
nm -- nanometers
SEM -- scanning electron microscope
std -- standard (state of gas)
|j,g -- micrograms
y,m -- micrometers
NOMENCLATURE
f -- fraction, PL/PT
F_ -- effective friction loss, cm WC
t,
N -- number of transfer units, dimensionless
p.. -- liquid feed pressure, atm gauge
Ap -- pressure drop, cm WC
3
P_ -- gas-phase contacting power, kWh/1000 m
3
P -- liquid-phase contacting power, kWh/1000 m
L 3
P -- total contacting power, kWh/1000 m
-------
Y rwjwinnn m3i "V
a -- coefficient of PT in equation N = o/PY, [kWh/1000 m3]
Y -- exponent of P in equation N = orP_ , dimensionless
T) -- fractional collection efficiency, dimensionless
1 - T] -- fractional penetration, dimensionless
xi
-------
ACKNOWLEDGMENT
The authors wish to express their appreciation for the cooperation
and encouragement of the Project Officers, Dr. Leslie E. Sparks and his
successor, Mr. Dale L. Harmon, of the Particulate Technology Section,
Industrial Environmental Research Laboratory, Environmental Protection
Agency. El Lorraine Watson performed most of the chemical analyses.
Stanford Research Institute supported the development and testing
of the ultrasonic aerosol generator used in this investigation.
xii
-------
SECTION 1
INTRODUCTION
GENERAL
Wet scrubbers are used widely for removal of particulate matter,
including submicrometer particles, from waste gases. The scrubbers in
use are able to attain high collection efficiencies if sufficient energy
is applied to the gas/liquid contacting process. However, the energy
required to attain a given efficiency increases with decrease in particle
size and becomes particularly high for particles smaller than one
micrometer.
There are three basic ways in which the necessary energy can be
introduced into the gas/liquid contacting process: from the gas stream,
from the liquid stream, or through a mechanical rotor. There are al-
most innumerable variations in the details of methods and equipment used
to introduce the energy by these paths. However, it has not been known
what method of introducing the energy produces optimum results, or even
whether there is a clear and definite optimum method. A considerable
body of data has indicated that the collection efficiency of a scrubber
is determined primarily by the power per unit of volumetric gas flow
rate ("contacting power") consumed in the gas/liquid contacting process,
and is not much affected by the method of introducing the power or by
the configuration of the device in which contacting takes place. The
equivalence of most scrubbers that take contacting power from the gas
stream in the form of pressure drop is quite well established by re-
sults of specific studies and by a body of general experience. Addi-
tional information has also indicated that contacting power taken from
the liquid stream is probably equivalent to that taken from the gas
stream; this indication has been modified by the results of the present
investigation. A small amount of data suggests that contacting power
from a mechanically driven rotor may also be equivalent, but that ques-
tion remains unresolved.
Obviously, it is extremely important for scrubber selection and de-
sign to know whether there actually are significant differences in the
contacting power/efficiency relationships corresponding to the different
ways of introducing the contacting power. Even if different scrubbers
showed no significant differences in inherent collection efficiency at
a given contacting power, specific scrubber configurations may offer
distinct practical advantages for particular applications.
-------
OBJECTIVES
The broad objectives of this investigation were:
(1) To determine the effect of the method of introducing con-
tacting power on the collection efficiency of particulate
scrubbers.
(2) To determine the effect of aerosol particle size on the
contacting power/efficiency relationship for particles in
the low-micrometer and submicrometer size range.
The specific tasks of the scope of work included the following:
(1) Review and assessment of the published literature and
available unpublished information pertaining to the effects
of energy consumption and different methods of energy
application on the collection of particles by scrubbers.
(2) A bench-scale experimental study of fine-particle collec-
tion by representative types of scrubbers that derive
contacting power from (a) the gas stream, (b) the liquid
stream, and (c) a mechanical rotor. Three test aerosols
were to be used, all having particle diameters of less
than 3 micrometers and at least one having a particle
diameter of less than 0.8 micrometer. The experiments
were to include study of series scrubbing through use of
multiple contactors or scrubbers in series.
(3) For all the scrubbers studied:
(a) Development of quantitative relationships between the
energy supplied to the scrubber and the particle col-
lection efficiency.
(b) Explanation, to the extent possible, of any differences
that might appear among the contacting power/collection
efficiency relationships for different types of
scrubbers.
(c) Identification, so far as possible, of the mechanisms
by which the energy supplied to the scrubber is used
to collect particles.
(d) Identification, so far as possible, of the energy-
consuming mechanisms.
(4) Recommendation of existing designs, design modifications,
operating practices, or othe,r factors in scrubber performance
that might be found to optimize particle collection effi-
ciency in existing scrubbers, especially for fine particles.
(5) Development of specific recommendations, together with cost
and time estimates, for a program to research, develop, and
demonstrate scrubbers that will make optimum use of energy
and liquid for particle collection.
-------
SECTION 2
SUMMARY AND CONCLUSIONS
The investigation of the energy consumption and collection efficiency
of scrubbers was accomplished by comparing the performances of different
devices on the same synthetic test aerosols. All tests were conducted
with ambient air as the gas and water at ambient temperature as the
scrubbing liquid. The performance of an orifice scrubber, a unit of
conventional design that had been studied earlier, was determined and
served as the baseline for comparison with other devices. The critical
element in the experimental program was ensuring that the test aerosols
remained consistent. Producing test aerosols of tolerable consistency,
determining that the consistency was maintained in each test run, and
determining the particle sizes and size distributions consumed by far
the greatest part of the project effort. By comparison, the remainder
of the experimental techniques used in scrubber testing were relatively
simple.
The aerosols were produced from sprays of ammonium fluorescein
solutions. Upon drying, the droplets left solid, spherical particles
of sizes determined by the original droplet sizes and by the concentra-
tion of ammonium fluorescein in the solution. The residual solid was
indicated to be the monoammonium salt, which has a very low solubility
in water and is nonhygroscopic. Very small quantities of the aerosol
could be determined precisely by fluorimetric analysis.
All the available practical methods for spray generation produced
polydisperse particles. Initially, an effort was made to use an SRI-
developed pneumatic atomizer, but this unit failed to give consistent
results. The effort was switched to a second SRI-developed aerosol
generator based on an ultrasonic nebulizer with numerous modifications
designed to enhance the stability. This unit still showed variations in
the aerosol produced, but in general the consistency was tolerable.
Monit'oring the consistency of the aerosol remained a problem, but was
accomplished by sampling the aerosol entering the scrubber with a "mini-
scrubber."
The miniscrubber consisted of four Greenburg-Smith impingers in
series, followed by a final filter. The unit was always operated at the
same gas pressure drop. By analysis of the aerosol collected in each
impinger and on the final filter, the collection efficiency of each stage
could be calculated. The efficiencies were expressed as the number of
transfer units, and the cumulative number of transfer units at each col-
lection stage was plotted as a function of the cumulative gas pressure
drop. The resulting curve, which was termed the "aerosol signature" or
-------
"signature curve," was characteristic of a given aerosol, and changes
in the curve denoted a change in the aerosol. If the degree of change
was excessive, the results of the test could be excluded.
The character and particle-size span of an aerosol were determined
from scanning electron microscope studies of the aerosol collected on a
Nuclepore membrane filter. The mass particle-size distributions and
mass-median diameters were determined from measurements with an Andersen
cascade Lmpactor with greased impact plates.
Three test aerosols were used. These were designated as D, E, and
G, and had indicated mass-median particle diameters of 1.05, 0.68, and
0.42 |jbm, respectively. The standard deviations of the distributions were
about 1.6-1.7.
The gas/liquid contactors investigated are described in summary in
Table 1. The orifice contactor was generally representative of conven-
tional orifice or venturi contactors. It had been extensively investi-
gated with a different aerosol in an earlier study, and was taken as the
reference unit. Performance curves (curves of number of transfer units
versus contacting power on a log-log scale) were determined for the
orifice scrubber for each of the three aerosols. All three aerosols
proved to have two-branched performance curves.
The multiple-orifice series contactor consisted of four staggered,
eccentric orifices in series. This unit gave the same performance as
the single-orifice scrubber over the upper part of the contacting power
range, but lower efficiencies over the lower part of the range. The
magnitude of the deviation and the contacting power level below which
the deviation appeared both increased as the aerosol particle size de-
creased. It is not known whether this behavior is characteristic of
series contactors in general, but there was no indication that series
contactors should be expected to give performance superior to that of a
single-stage contactor.
With the spray contactors and the combination spray-orifice con-
tactor, some friction loss (and gas pressure drop) always attended flow
of the air through the scrubber, so that the total contacting power al-
ways included gas-phase as well as liquid-phase components.
The SS5 and SO configurations could be operated as ejectors, with
all the draft for air flow through the scrubber being supplied by the
water jets from the nozzles.
Except when the relative amount of liquid-phase contacting power
was very low, the efficiencies of the spray scrubbers and the spray-
orifice scrubber were definitely lower than those of the orifice scrubber.
The deviation from the orifice scrubber performance curve proved to be
correlated by a simple parameter _f, the ratio of the liquid-phase con-
tacting power to the total contacting power. The deviation increased
with increase in the value of _f. However, for given values of _f, spray
-------
TABLE 1. GAS/LIQUID CONTACTORS
Contactor
Orifice
Multiple orifices
in series
Spray
Spray
Spray
Spray
Spray
Spray and orifice
Test
code
OS
MOS
SSI
SS2
SS3
SS4
SS5
SO
Gas line
diameter
(cm)
5.84
5.84
5.84
5.84
5.84
5.84
5.84
5.84
Orifice
diameter
(cm)
2.54
3.81
3.36a
2.54
3.81
Spray nozzle
type
Perforated tube
(multiple per-
forations)
Flat spray
Full cone,
wide angle
Deflected,
hollow cone
Full cone, in-
jector type
Full cone
Nozzle position
in gas line
Concentric
Sidewall
Concentric
Concentric
Concentric
Concentric
Spray orientation
Radial, crossflow
Crossf low
Cocurrent
Radial, crossflow
Cocurrent
Cocurrent
Spray
angle
0°
95°
100°
180°
30°
60°
Diameter of circle having same area as semicircular orifice.
-------
scrubber configurations SSI, 2, and 3 gave almost identical performances
despite radical differences in nozzle design and orientation and in spray
characteristics. Configurations SS5 and SO gave only slightly better
performance than SSI, 2, and 3. Only the SS4 configuration gave an
efficiency markedly different (lower) from the general level of the
other spray contactors.
When configurations SS5 and SO were operated as ejector scrubbers,
the performance at a given contacting power agreed well with that of
other spray configurations at high values of _f.
Because of the large effort that had to be devoted to aerosol de-
velopnent, measurement, and monitoring, the time and funds available for
the project were not sufficient to permit an investigation of mechanical
rotor gas/liquid contactors.
Some major conclusions from the investigation are:
(1) No indication was found of the existence of a narrowly de-
fined optimum way of applying contacting power in particu-
late scrubbing. Further evidence was added to earlier
observations that for a given aerosol there is a level of
performance that cannot be exceeded with a given gas-
phase contacting power; certain scrubber configurations or
operating modes may give poorer performance than the "norm,"
but not better performance.
(2) Liquid-phase contacting power was found to be broadly less
effective than gas-phase contacting power. However, it is
not clear whether the apparent inferiority of the liquid-
phase contacting power is real, or whether the discrepancy
may be partly or wholly due to the difficulty of estimating
the quantity of liquid-phase contacting power actually con-
sumed in gas/liquid contacting, which cannot be measured
directly.
(3) The relative contribution of liquid-phase contacting power
to collection efficiency was not increased by closely com-
bining the action of the spray with that of a gas-phase
turbulence promoter, as in the spray-orifice combination
contactor.
(4) Although the various spray-type gas/liquid contactors pre-
sumably yielded sprays of radically different characters,
these differences produced no evident substantial effects
on collection efficiency (except possibly in one case)
that were independent of the contacting power applied.
(5) The energy consumed in scrubbing is dissipated in fluid
friction and turbulence and is the same whether or not
particles are present in the gas phase. The energy dis-
sipation is associated with conditions that lead to particle
collection, but the collection process does not involve
application of energy directly to the particles.
-------
(6) The relationships observed between scrubber contacting power
and collection efficiency do not appear currently to be
either predictable or explicable from first principles.
-------
SECTION 3
RECOMMENDATIONS
Since the maximum attainable level of particulate scrubber collec-
tion efficiency is evidently determined essentially by the contacting
power level, no recommendations can be offered for optimizing the per-
formance of either existing or potential new designs. However, since
certain scrubber configurations or operating modes have been found to
deliver less than the maximum level of performance, it is highly de-
sirable to avoid using such configurations unless they have other charac-
teristics favorable for specific applications that will outweigh their
deficiencies in contacting power/efficiency ratio.
The: comparative performance of mechanical rotor scrubbers should
also be determined. Although there is currently no reason to expect
that the mechanical scrubbers will give performance significantly
superior to that of venturi or orifice scrubbers at the same contacting
power, they might have practical advantages for certain applications.
For example, in an application requiring very high contacting power,
the use of a mechanical contactor will avoid the necessity of moving
the gas through very large pressure differentials using high-pressure
blowers.,
The type of experimental approach used in the present investiga-
tion, bunch-scale comparative testing of basic devices with standardized
aerosols over wide ranges of conditions, is a practical one that should
be applied more widely. It is essentially a simple technique, provided
that a consistent aerosol is available.
-------
SECTION 4
BACKGROUND
Although some observers had previously noted that high collection
efficiency in particulate scrubbers was generally associated with high
power consumption, it was the work of Lapple and Kamack1 that first
showed that, for a given aerosol, a functional relationship exists be-
tween the collection efficiency and the gas pressure drop of scrubbers
in which the energy consumed is derived from the gas stream. Equally
striking was the observation that different designs of gas/liquid con-
tactors show only modest differences in collection efficiency when
operated at the same gas pressure drop. Lapple and Kamack suggested
that power input from spray nozzles and mechanically driven rotors might
also correlate with scrubber efficiency. Additional evidence for the
relationship of scrubber efficiency and pressure drop was obtained from
proprietary studies at Stanford Research Institute.2 Still further
support was obtained by Semrau et al.3 in field studies of pilot plant
scrubbers; this work also showed considerable support for the suggestion
that power input from spray nozzles might be equivalent to that from gas
pressure drop.
The concept of "contacting power" was extended and refined in later
publications by Semrau4'5 (see Appendix A). Correlations of published
data from various laboratory and field studies gave additional indica-
tions that contacting power from spray nozzles and mechanical rotors
might be equivalent to that from the gas stream. Unfortunately, no
comparable data taken under carefully controlled conditions have been
available to resolve firmly the questions concerning the relative effec-
tiveness of gas-phase, liquid-phase, and mechanical contacting power.
A variety of studies by other investigators has generally supported
the equivalence of gas-phase contacting power in different designs of
gas/liquid contactors, though still other workers have reported substan-
tial deviations (see Appendix A). Although claims of greatly superior
performance are still sometimes made for certain devices, based on
supposed novelties of scrubber design or operation, none of these claims
have so far been sustained when subjected to critical experimental
testing. In a recent study with an orifice scrubber with a synthetic
ammonium fluorescein aerosol, Semrau and Witham6' 7 found the collection
efficiency to be a function only of contacting power, not independently
affected by orifice size, gas velocity through the orifice, or liquid-
to-gas ratio. These results supported earlier findings made with much
more restricted ranges for the variables.
Since 1963, basic studies of scrubbers have largely been concerned
with efforts to develop theoretical models of scrubber operation from
-------
first principles. The most elaborate of these studies have been devoted
to ;he venturi scrubber; among the better known models that have been
developed are those of Calvert,8""10 Boll,11 and Behie and Beeckmans.12>13
Al?. these models are based on collection of particles of discrete sizes.
Calvert has also modified his models to express the scrubber collection
efficiencies in terms of the gas pressure drops. However, Calvert's
equations contain a factor that absorbs errors and approximations in the
derivation of the models, and this factor had to be evaluated from ex-
perimental data.
Unfortunately, there are very few experimental data that could be
used to evaluate the factor in the Calvert equations or to test the
theoretical models of Calvert or other workers. Some investigators have
noted substantial deviations between the predicted scrubber performance
and that indicated by available experimental data.11'13 Attempts at
experimental verification have commonly covered only narrow ranges of
operating variables. Testing of the models requires use of monodisperse
aerosols or of measurements of the collection efficiency for particles
of discrete sizes if polydisperse aerosols are used. The production of
monodisperse aerosols is difficult to accomplish and control, and a
narrow size distribution rather than true monodispersivity is the best
that can be obtained in practice. Measurements of particle-size dis-
tribution at the inlet and outlet of a scrubber are also difficult and,
in seme circumstances at least, of questionable reliability. The sheer
laboriousness of the latter technique, in particular, usually limits the
number o:3 determinations and the ranges of scrubber operating variables
that can be reasonably investigated.
The experimental approach used in the present study, comparative
testing of scrubber units with standardized aerosols, is a relatively
simple technique, although producing and monitoring a "standardized
aerosol" is by no means simple, as is demonstrated below. Ideally, the
test aerosol should be monodisperse, to permit evaluation of the scrubber
performance in terms of absolute particle size. But even if production
of a nonodisperse aerosol is impractical, it is possible to make reliable
comparisons of the actual behavior of different scrubbers. Essentially,
the scheme is to reduce all determinations to those quantities that can
be measured with relative simplicity and confidence. As far as possible,
it is necessary to avoid making the comparison of devices dependent on a
chain of measurements that may accumulate errors or uncertainties.
The experimental problem lies primarily in generation of the aerosol,
which oiust be stable, reproducible, and readily measurable in quantity
if not in particle size. In addition, the generation technique must be
rapid and relatively simple to permit large numbers of tests. It is far
more critical that the aerosol particle-size characteristics remain
constant than that they be known in detail on an absolute basis. Never-
theless, it is important to know as much as practical about the particle-
size characteristics of the aerosol. In particular, the aerosol should
be dispersed as discrete particles to the greatest possible extent. In
practice, this requires use of low aerosol concentrations to limit
10
-------
coagulation, which in turn requires choice of an aerosol material that
is readily detenninable in small quantities.
The nature of an aerosol material and the available technique for
generation are interrelated. Attaining given desired aerosol properties
commonly requires sacrificing others, so that any choice represents a
compromise.
11
-------
SECTION 5
EXPERIMENTAL METHODS AND EQUIPMENT
PILOT-PLANT SCRUBBER
The small pilot-plant experimental scrubber system is portrayed
schematically in Figure 1. The same unit was previously used in a
project entitled "Wet Scrubber Liquid Utilization" (EPA Contract No.
68-02-1079) , which was actually concerned with condensation and evapora-
tion effects in scrubbing and was reported by Semrau and Witham.6 The
components of the original system that were not used in the present
study have been omitted from Figure 1. This system was designed to
accommocate inlet gas flows of up to about 4200 liters/rain (150 ft-Vmin)
at ambient temperature, and with gas pressure drops across the scrubber
of up to 300 cm WC (120 inches WC). The scrubber itself consisted of
the contactor section followed by a cyclone entrainment separator. The
contactor section consisted of a vertical section of line in which the
chosan gas/liquid contactor was mounted and through which the gas flowed
downward. The gas and water leaving the contactor section passed
through a mitre elbow before flowing horizontally into the cyclone
separator.
When an orifice gas/liquid contactor was used, the orifice plate
was mounted in a horizontal plane between flanges in the vertical line.
Interchangeable orifice plates were used to vary the diameter of the
contacting orifice. The scrubbing water was introduced through a single
tap in >;he wall of the inlet line, 2 pipe diameters above the orifice
plate. The orifice scrubber design thus followed on a small scale the
general configuration currently being used for most commercial scrubbers
of the orifice or modified venturi types.
Air was drawn through the system by a rotary positive-displacement
blower of the Roots type. The main air flow through the system was drawn
directly from the room and metered before entering the scrubber. The
air flow through the system was controlled by bleeding air from the room
directly to the inlet of the positive-displacement blower. The air by-
pass; intake was fitted with a filter-silencer and the blower with a dis-
charge silencer. However, the main air stream entering the system was
not filtered; since the test aerosol was measured by a specific fluori-
metric method, the presence of atmospheric dust in the room air did not
interfere with the determination of scrubber efficiency.
The contactor section was made from 2-1/2-inch stainless steel
tubing (5.84 cm inside diameter). The cyclone separator, which was
actually larger than necessary to handle the gas flows used, was 45.7 cm
12
-------
BURNER
Room
Air
Main
Air -
Flow
V
Drying
Air
*** *~
Jatural
Gas
1
/
\
\\
f
\ 1
\
~-
L
-,
:ROSC
kEROSOL I
AND
Inlet
Sample
1.
MANOMETERS
(jM 1 \
(jM
3RIER
ft/ii y pp
ivi i A c n
— CT0
[ TI 1
1 I 1 1
r
r*"i GAS/LIQUID
1 CONTACTOR VlX '^
(£) ! SECTION By - Pass
1 r— 1 1 * * M- Ajr
1 1 /^N Outlet
J V } Sample
\/ CYCLONE
/ PMTOAIMMCMT* -T« C»«.l,
/ SEPARATOR ( T )
DriTAIWICTCO / N S*-—^\
1
Aerosol Sweeo
Air
1
1
)L —
-1 BLOWER
o r
Xx
Water
PUMP TO
Drain
Fresh
GENERATOR ~| Water
Compressed
Air
Compressed
Air SA-438O-5O
FIGURE 1 FLOWSHEET OF EXPERIMENTAL SCRUBBER SYSTEM
-------
(18 inches) in diameter, with an inlet 7.6 cm (3 inches) in diameter and
an outlet 10.2 cm (4 inches) in diameter. The general design was the
same, as that used by Lapple and Kamack1 for a semiworks scrubber unit.
The main inlet air header, including the orifice meter section, was made
from 6-inch (15.2-cm) stainless steel tubing.
The main air flow was measured with a standard-design orifice meter
equipped with square-edged orifices and fitted with radius taps located
1 pipe diameter upstream and 1/2 pipe diameter downstream of the orifice
plate. Orifice meter differentials and air system pressures and pressure
drops ware measured with U-tube manometers filled with xylene or mercury,
as demanded by the ranges of pressures involved. The air temperatures
were measured with dial thermometers of the bimetal type.
Water was supplied to the gas/liquid contactor by one of two pumps,
one a centrifugal for low-pressure service and the other a turbine for
high-pressure service. The water was supplied from the building supply
line at ambient temperature. The water flows to the scrubber were
measured with rotameters of appropriate ranges. The water pressure up-
stream of the spray nozzles used as gas/liquid contactors was measured
with a precision Bourdon-type gauge.
The scrubber and cyclone were elevated so that the water could be
discharged by gravity through a seal loop at the highest gas pressure
drops used in the tests.
GAS/LIQUID CONTACTORS
Tho essential characteristics of the gas/liquid contactors studied
are summarized in Table 1. The orifice and multiple-orifice series con-
tactors are shown in Figures 2 and 3. The series contactor was formed
by mounting half-discs in a staggered array within a tube, which was
flanged at one end. The tube was mounted concentric with the contactor
line and retained with the flange held between the orifice flanges.
With both these contactors, the water was injected under negligible
pressure through the large (1/2-inch IPS) feed tap.
All the spray nozzles except the perforated tube (SSI) were commer-
cial models produced by Spraying Systems Company. The manufacturer's
designations are presented in Table 2. The methods of mounting the
nozzles are illustrated in Figures 4-9.
AEROSOL GENERATION
The aerosol generation system is indicated schematically in Figure 1.
The aerosol generator produced a cloud of droplets of ammonium fluorescein
solution. A small flow of air swept the mist out of the generator and
straight upward through a section of 1-inch IPS pipe. The 1-inch pipe
extended to the lower end of a conical diffuser section, which was located
14
-------
GAS
FLOW
WATER
INJECTION
ORIFICE
FLANGES
SA-4380-42
FIGURE 2 ORIFICE GAS/LIQUID CONTACTOR
15
-------
GAS
FLOW
SEMICIRCULAR
ORIFICE
WATER
INJECTION
ORIFICE
FLANGES
SA-4380-41
FIGURE 3 MULTIPLE-ORIFICE SERIES GAS/LIQUID CONTACTOR
16
-------
TABLE 2. PRESSURE SPRAY NOZZLES USED FOR GAS/LIQUID CONTACTING
SRI
test
code
SSI
SS2
SS3
SS4
SS5
SO
Nozzle
manufacturer
SRI
Spraying
Systems Co.
Manufacturer ' s
designation
Perforated tube
H 1/4 U9530 VeeJet
3/8 G17W FullJet
8686-1/4-1.5-180°
DeflectoJet
1/2 G3030 FullJet
3/8 G15 FullJet
Nozzle type
Multiple solid-stream
water jets
Flat spray pattern
Full-cone wide angle
spray pattern
Deflected type hollow-
cone spray pattern
Full-cone spray pattern,
30°, injector type
Full-cone spray pattern
-------
WATER
INJECTION
SA -4380-43
FIGURE 4 SPRAY GAS/LIQUID CONTACTOR
Configuration SS1
18
-------
GAS
FLOW
WATER
INJECTION
SA-4380-44
FIGURE 5 SPRAY GAS/LIQUID CONTACTOR
Configuration SS2
19
-------
WATER
INJECTION
100°
SA-4380-45
FIGURE 6 SPRAY GAS/LIQUID CONTACTOR
Configuration SS3
20
-------
WATER
INJECTION
AIR
FLOW
SA-4380-46
FIGURE 7 SPRAY GAS/LIQUID CONTACTOR
\
Configuration SS4
21
-------
WATER
INJECTION
SA-4380-47
FIGURE 8 SPRAY GAS/LIQUID CONTACTOR
Configuration SS5
22
-------
WATER
INJECTION
SA-4380-48
FIGURE 9 SPRAY-ORIFICE GAS/LIQUID
CONTACTOR
23
-------
concentrically within the 15.2-cm (6-inch) riser that carried the main
air flow upward to the scrubber. Another small flow of drying air also
entered the lower end of the conical diffuser section through the annular
space around the 1-inch pipe carrying the mist. The mist and the drying
air mixed in the diffuser section, where the water was evaporated from
the droplets, leaving solid aerosol particles. The resulting aerosol
stream emerging from the end of the diffuser then mixed with the main
air stream, which flowed through the annular space between the upper
end cf the diffuser and the 15.2-cm riser.
The sweep air that carried the aerosol out of the nebulizer was
metered with a critical-flow orifice. The flow rate used was 6 liters/min
with the coarsest aerosol and 2.1 liters/min with the two finer aerosols.
The aerosol drying air was also metered with a critical-flow orifice to
a constant rate of 84.9 liters/min (3 ft^/min). The drying air stream
coulo. be heated with a small direct-fired gas burner if necessary. How-
ever, ths dry air was found to be capable of evaporating the water from
the c.roplets without being heated above room temperature.
The aerosol generator itself was developed by Stanford Research
Institute, and was based on use of a commercial ultrasonic nebulizer to
produce the mist. However, an elaborate arrangement of auxiliary equip-
ment was necessary to convert the nebulizer into an aerosol generator
suitable for the present study. The resulting unit was vastly superior
to an earlier ultrasonic generator used in the previous study.6 The
generator is described in more detail in Appendix B.
The ammonium fluorescein used for the aerosols had already proved
highly satisfactory in the previous study.6 It was desired to produce
three simlar aerosols having nominal mass-median diameters of 0.4, 0.7,
and 1.5 ;j,m. The generator produces droplets with a nominal mean size of
about. 6-8 n,m. It was thereby determined that the three aerosols could
be produced from ammonium fluorescein solutions having concentrations of
0.01%, 0.1%, and 1.0% by weight. The ammonium fluorescein was first pre-
pared at a concentration of 17., by weight. The appropriate amount of
fluorescein (acid form) was weighed out and dissolved in a quantity of
ammonium hydroxide solution equivalent to four moles of ammonium ion per
mole of fluorescein (acid form). Water was added to make up the solu-
tion to the appropriate total weight (and volume).
In the first phase of the project, the 0.1% and 0.01% solutions
were prepared by diluting the 1.0% solution 10 to 1 and 100 to 1, re-
spectively, with 0.3N ammonium hydroxide. It was then discovered that
generation of the aerosol from the 0.01%, solution was being affected by
contiiminants--apparently silica dissolved in the ammonium hydroxide
solution. Thereafter, the 0.1% and 0.017o solutions were prepared by
diluting the 17» stock solution with distilled water instead of 0.3N
ammonium hydroxide. The behavior of the generator and aerosols is dis-
cussed in Section 6.
24
-------
AEROSOL SAMPLING AND CHARACTERIZATION
The aerosol was sampled at the inlet and the outlet of the scrubber
(see Figure 1). At the outlet, the sample was collected on a 25-mm
Nuclepore membrane filter (0.6-|jim pore size) held in a Nuclepore holder
that was mounted in the scrubber outlet line. The inlet piece of the
filter holder was machined out to expose almost the entire effective
area of the membrane filter directly to the incoming gas, and the holder
was turned to face directly into the gas stream being sampled. At the
inlet of the scrubber, the aerosol sample was obtained with the mini-
scrubber (described below), and the air sample was drawn out of the duct
through a sampling nozzle that consisted of a 90° ell of 2.54-cm stain-
less steel tubing. The open end of the ell faced directly into the gas
stream. No effort was made to sample isokinetically at either inlet or
outlet because the aerosols were too small for sampling to be influenced
appreciably by inertial effects. The air flows through the sampling
trains were metered with critical-flow orifices. Vacuum was supplied by
a mechanical vacuum pump. The gas sampling rates were 16 liters/min at
the scrubber inlet and 6.5-7.5 liters/min at the outlet.
The sampling techniques used in this investigation were essentially
the same as those used in the earlier study,4 in which it was determined
that the average error in aerosol sampling did not exceed 57o-67<>. No
additional determinations of sampling errors were made in this investi-
gation.
The miniscrubber consisted of four nominally identical Greenburg-
Smith impingers (Ace Glass Inc. No. 7536) connected in series. A
diaphragm-type differential pressure gauge (Dwyer Magnehelic) was used
to measure the total pressure drop across the four impingers, which was
standardized at 200 cm WC. Because of minor differences in the impinger
nozzles and/or impact plates, the pressure drops of individual impingers
showed variations of up to about 1470. These variations were taken into
account in the correlation of miniscrubber sample data. Identical
quantities of liquid (100 ml of 0.3N ammonium hydroxide solution) were
used in each impinger. The ammonium hydroxide solution was used to
ensure that the collected aerosol particles would go into solution,
since the monoammonium fluorescein aerosol material has a very low
solubility in pure water.
A 47-mm Nuclepore filter (0.4-|o,m pore size) was used to collect the
aerosol that penetrated the impinger train. From analyses of the aerosol
collected in each impinger and on the filter, it was possible to deter-
mine the collection at each stage of the miniscrubber as well as the
overall efficiency. The methods of presenting and using the miniscrubber
test data are described and discussed in Section 6.
Aerosol particle size was studied by two methods. Direct visual
indications of aerosol size and character were obtained with a scanning
electron microscope (SEM). Samples were collected on Nuclepore membrane
filters having either 0.2- or 0.4-^m pores. A sample was then prepared
for study by cutting a section out of the center portion of a membrane
25
-------
and mounting it on an aluminum SEM stage, fixing it in position with
conductive silver paint. The prepared sample was coated with a thin
film of gold-palladium by vapor deposition to render the surface con-
ductive. The sample was scanned with the SEM to locate one or more
representative fields. Then photomicrographs were made at a number of
magnifications. From these it was possible to estimate the range of
particle sizes. It was also possible to make a rough visual judgment of
the probable mass-median diameter, but no attempt was made to determine
the distributions by making particle counts.
Measurements of the particle-size distributions of the aerosols
were made with an eight-stage Andersen nonviable cascade impactor (manu-
factured by Andersen 2000 Inc.) equipped with glass impaction plates.
The impactor was followed by a 47-mm Nuclepore filter with 0.4-|j,m pore
size. The air flow rate was measured with a Sprague dry gas meter and
was controlled with a valve in the vacuum line downstream of the im-
pactcr, backup filter, and meter. Air flow rates were set at the be-
ginning of a run and held constant throughout the run. The rates varied
from about 15 to 25 liters/min and produced a total pressure drop across
the impactor and the backup filter of 1-15 cm of mercury. The calculated
flow rates were corrected for the temperature and pressure existing at
the meter.
The initial impactor studies were made with an older seven-stage
Andersen unit, using dry glass impaction plates. The early results in-
dicated that the aerosols were finer than would be anticipated from the
SEM examinations, suggesting the possibility that bouncing and reentrain-
ment of the aerosol particles may have been taking place. Later, when
the r.ew aight-stage impactor was substituted, the collection plates were
coated with Dow-Corning silicone stopcock grease. The coating was
applied 'Dy painting the plates with a 1070 solution of the grease in
carbon tatrachloride, using a soft camel's hair brush. Two applications
were mada, with an intervening drying period, and the plates were
allowed to dry under a clean hood. As observed under a microscope, the
dried grease surfaces were not smooth; however, no attempt was made to
alter tha surface after the grease was applied.
Trials were also made with the impaction plates covered with sticky-
faced paper. However, this method was abandoned as inferior to use of
the greased plates.
AEROSOL ANALYSIS
Fluorimetric analyses of the ammonium fluorescein aerosol material
were made with an Aminco-Bowman Spectrophotofluorometer with a xenon
lamp (Catalog Numbers 4-8202 and 4-8202B). The unit was equipped with
monochro'iiators for both excitation and emitted light. Instrument settings
used in the analyses were as follows:
26
-------
• Excitation wave length--482 rim.
• Emission wave length--540 nm.
• Primary slits--full open.
• Secondary slits--5 mm.
• Photomultiplier slit--0.1 mm.
• Meter multiplier setting—to give transmittance reading
between 30% and 90%.
A Corning 3-69 filter was used to eliminate interferences with excitation
wave length.
The transmittance of the ammonium fluorescein solutions followed
Beer's Law at concentrations between 4 X 10"? g/cm^ and 1 X 10~1° g/cm^.
The fluorescence was markedly a function of pH, increasing by a factor
of 25 as the pH was increased from 4 to 10. Samples were dissolved in
a solution of ammonium hydroxide having an initial pH of 11.
Samples for analysis (1.5-2 cm3) were placed in a clean, dry-cuvette
(fused quartz) of rectangular cross section. If the resulting concen-
tration of ammonium fluorescein was above the preferred range for
analysis, the sample was diluted with an ammonium hydroxide solution of
pH 11.
Since the liquid from the impingers was a 0.3N solution of ammonium
hydroxide (pH 11), it could be analyzed directly. All the internal
glass surfaces of an impinger were washed, and the washings were included
with the liquid from the impinger. A single washing of a Nuclepore
filter was sufficient to remove the aerosol deposit; therefore, filters
were simply immersed in 10 cm^ of the ammonium hydroxide solution and
agitated.
With the Andersen impactor, each orifice plate and each dry collec-
tion plate was washed in 20 cm^ of the ammonium hydroxide solution. The
orifice plates were washed and the washings analyzed to determine wall
losses.
When grease-coated impaction plates were used, a plate was washed
with carbon tetrachloride and agitated until no remaining grease could
be seen on the plate. A portion of ammonium hydroxide was then added
to the carbon tetrachloride washings and the two fluids were agitated
together. The ammonium fluorescein is essentially insoluble in the carbon
tetrachloride. Since the carbon tetrachloride and the ammonium hydroxide
are immiscible, the ammonium fluorescein was extracted from the carbon
tetrachloride into the aqueous solution phase, which was then separated
and analyzed.
Experience with replicated analyses indicated that the error in
ammonium fluorescein analysis did not exceed 17o except in the very
smallest samples, for which the error was somewhat larger.
27
-------
GENERAL SCRUBBER TEST PROCEDURE
All tests with the scrubber were made with air and water at ambient
temperatures. The average room air temperature was about 25°C. When
the scrubber operating conditions were adjusted as desired and operation
of the sctubber and aerosol generator was steady, the samplers were
operated for a measured time interval (generally 5-20 minutes, depending
on the aerosol used). All air flow rates in the scrubber and the samplers
were referred to the chosen standard state, 25°C and 1 atm pressure,
which was approximately the average condition in the laboratory.
28
-------
SECTION 6
RESULTS AND DISCUSSION
AEROSOL CHARACTERIZATION WITH THE MINISCRUBBER
Since the basic experimental approach was absolutely dependent on
the ability to produce consistent and reproducible test aerosols, it was
necessary to have an appropriate method for determining the consistency
and reproducibility of the aerosols being generated. The available
methods for particle-size analysis are generally slow or complicated, or
both, and are subject to a variety of errors and uncertainties. Also, a
given method is commonly not applicable over the complete spectrum of
particle sizes that may be encountered. For the present study, it was
essential that the method for monitoring the consistency of the aerosols
should be simple, rapid, and reliable. Fortunately, it was not necessary
that the monitoring method give an absolute measurement of particle-size
distribution; it was sufficient that the technique reveal changes in the
aerosols and give some idea of the magnitude of the changes.
In previous investigations, it was observed that with scrubbers
using gas-phase contacting power, the performance of the scrubber at a
given contacting power or, better, over a range of contacting power was
a very sensitive indicator of differences in aerosols. The contacting
power/efficiency relationship does not, of course, provide an absolute
measurement of particle size or particle-size distribution, although it
does give a relative indication of particle size. The use of the mini-
scrubber was not predicated on any assumption that the relationship of
collection efficiency to pressure drop should necessarily be the same as
that for larger scrubbers of the same or different types. The only
assumption was that if the miniscrubber was always operated identically,
it should always give the same efficiency on the same aerosol.
The use of multiple stages improved the aerosol characterization.
By expressing the stage efficiencies in terms of the number of transfer
units, which are additive, and plotting the cumulative number of transfer
units as a function of the cumulative gas pressure drop, a complete
scrubber performance curve could be generated from one test run. The
results were equivalent to four scrubber efficiency determinations at
different pressure drops obtained by using combinations of one, two,
three, and four scrubber units (impingers) in series.
As is shown below, the four data points commonly defined with high
precision a straight line on a log-log plot. The number of transfer
units obtained from a stage is qualitatively indicative of the particle
size of the aerosol entering that stage. The slope of the curve should
logically be related to the particle-size distribution. For an absolutely
29
-------
monodisperse aerosol, each identical scrubber stage should give the same
collection efficiency, and the slope of the performance curve should be
unity. This conclusion further assumes, however, that the aerosol is
not hygroscopic and does not grow between stages by absorption of moisture,
and that, the temperature and other conditions of the gas stream remain
constant throughout the scrubber train. If the aerosol has a distribution
of particle sizes, the preferential collection of the larger particles
will cause a reduction in the average particle size of the aerosol entering
each succeeding stage, and hence a drop in the collection efficiency of
each succeeding stage; this is the behavior actually encountered with the
aerosols generated. In such a case, if the performance curve is a
straight line (as has been the case in practice), the slope will be less
than unity.
In early work,5 it was hypothesized that the slope of scrubber per-
formance curves in general should probably be determined by particle-size
distribution. However, the results of work both at SRI and elsewhere5
have shown enough anomalies to make it highly unlikely that the slope is
determined only by the particle-size distribution, although this may be
more nearly true of the miniscrubber performance curves.
To avoid confusion in references to the performance curves of the
pilot-plant scrubber and the miniscrubber, it was desirable that the
curve for the miniscrubber be designated by another term. Since the
miniscrubber performance curve is presumably characteristic only of the
aerosol, the curve was termed the "aerosol signature curve," or simply
as the "aerosol signature."
TEST AEROSOL GENERATION
The ultrasonic aerosol generator was first tested extensively to
determine the stability and reproducibility of the aerosol production.
This part of the work was supported by SRl's Internal Research and
Development funds. Tests were made to determine the degree of inter-
changeability of different generator crystals and power sources. All the
tests were made with 1.0% ammonium fluorescein solution, and the repro-
ducibility of the aerosol was judged from the aerosol signature curves.
The tests and results are discussed in detail in Appendix B. In brief,
it was found that different generator crystals and power sources were
interchangeable, and that those originally put in use could therefore be
replaced if they should fail during the experimental program. The aerosol
produced did vary, but within a range that appeared likely to be tolerable.
The remainder of the characterization of the aerosol (designated as
Aerosol D) was carried out in conjunction with determination of the base-
line performance curves of the orifice scrubber. It was then found that
the performance of the miniscrubber was more sensitive to variations in
the aerosol than was the orifice scrubber, which provided some margin of
safety in the experimental program.
30
-------
Originally, it had been hoped that the aerosol consistency might be
sufficient that only periodic checking with the miniscrubber would be
necessary, and that the aerosol concentration at the bench-scale scrubber
inlet could normally be determined with a filter only. However, experi-
ence demonstrated that full confidence in the results required obtaining
the signature of the aerosol for each scrubber efficiency test. There-
after, the scrubber inlet sample was taken with the miniscrubber in all
cases. This procedure greatly increased the number of samples that had
to be prepared and analyzed.
The nominal mean diameters of the test aerosols were chosen to give
a factor of approximately 2 between the sizes of succeeding aerosols
(see Table 3). Aerosol E was similar in size to the aerosol used in the
previous study of condensation phenomena in scrubbing.6 Preliminary
verification of the approximate particle sizes of the aerosols was made
from studies of samples with the SEM. Because the particles were spheri-
cal (except as noted below) and of relatively low density (1.35 g/cnH),
there was reasonable agreement between the actual diameters as observed
in the electron micrographs and the aerodynamic diameters as determined
with the cascade impactor. The density of the monoammonium fluorescein
was not measured but was taken from a paper by Stb'ber and Flachsbart.14
The larger aerosols, D and E, were composed entirely of spheres (see
Figures 10 and 11). However, the first nominal 0.4-|J-m aerosol produced,
Aerosol F, was composed of irregular and flattened particles (see
Figure 12), so that the equivalent diameters could not be judged by
visual examination. Aerosol F was also far more erratic than Aerosols D
and E, as was shown by the signatures and the scatter in the orifice
scrubber performance data (discussed below). Upon study, it became
evident that contamination of the 0.017o ammonium fluorescein solution
being nebulized was responsible. Trial generation of aerosols was carried
out with solutions in the concentration range from 0.1% to 0.01%, and it
was found that the departure of the particles from spherical shape was a
trend appearing as the concentration was reduced below 0.057o. The most
likely contaminant appeared to be silica dissolved from glass containers
by the ammonium hydroxide solution. Because the 0.017<> solution was pro-
duced by dilution of the 1% stock solution with 0.3N ammonium hydroxide,
the ratio of ammonium hydroxide--and presumably of any dissolved silica--
to ammonium fluorescein was much greater in the dilute solution.
Some confirmation of this hypothesis was obtained from X-ray disper-
sion studies of the aerosol samples being examined in the SEM. Qualita-
tively, Aerosol F was found to have a much higher silica content than did
Aerosol D. In addition, a test was made by nebulizing 0.3N ammonium
hydroxide solution in the aerosol generator and sampling the air stream
with a Nuclepore filter for a prolonged period. Examination of the
filter sample with the SEM showed irregular particles that may have con-
sisted of ammonium silicate or silica. From examination of comparable
samples with the SEM, it was crudely estimated that the mass of the con-
taminant material present in Aerosol F might have been of the order of
one-tenth that of the ammonium fluorescein. Such a quantity of silica
(or even two or three times that amount) would have relatively little
31
-------
TABLE 3. TEST AEROSOL CHARACTERISTICS
NJ
Code
designation
D
E
F
G
Nominal
mean diameter
(um)
1.5
0.7
0.4
0.4
Particle-size
rangea
(um)
0.2 - 5
0.15 - 2.6
0.1 - 2.3d
0.07 - 1.3
Mass -median
diameter^5
(um)
1.05
0.68
0.42
Standard
geometric
deviationc
1.58
1.69
Observed from electron micrographs.
Measured with cascade impactor.
Q
Particle-size distribution taken as log-normal.
Particles flattened.
-------
LOW
MAGNIFICATION
1.0
2.0 urn
5.0
HIGH
MAGNIFICATION
0.4
1.0 Aim
2.0
SA-4380-57
FIGURE 10 ELECTRON MICROGRAPH OF AEROSOL D
33
-------
LOW
MAGNIFICATION
0.5
1.0
2.5
HIGH
MAGNIFICATION
0.4
1.0 /jm
2.0
SA-4380-58
FIGURE 11 ELECTRON MICROGRAPH OF AEROSOL E
-------
LOW
MAGNIFICATION
0.2
0.4 jum
1.0
HIGH
MAGNIFICATION
0.2
0.5 Mm
1.0
SA-4380-59
FIGURE 12 ELECTRON MICROGRAPH OF AEROSOL F
J5
-------
influence on the size of the aerosol particles, but it evidently did
have a doninant effect in determining the particle shape.
When the 0.017,, ammonium fluorescein solution was prepared by diluting
the 17o stock solution with distilled water instead of ammonium hydroxide
solution, the particles generated from it were generally spherical
(Figure l.'J). Some particles of irregular shape were still produced,
particularly those of the finest sizes that would presumably be most
affected by the presence of the contaminant. Probably the residual con-
tamination originated from the ammonium hydroxide solution used in pre-
paring the ammonium fluorescein from the acid form of fluorescein. The
new aerosol, designated as Aerosol G, was generally satisfactory and was
used in the subsequent studies. However, Aerosol G did show greater
variability than Aerosols D and E, as was shown by variations in the
signatures and scatter in the scrubber performance data. It appears
that in any future work the ammonium fluorescein should be prepared from
ammonia gas without any contact with glass.
Cther investigators who have used ammonium fluorescein have also
experienced contamination problems with solutions below 0.057o in concen-
tration. lli
The particle-size distributions of Aerosols D, E, and G, as deter-
mined from cascade impactor measurements, are shown in Figures 14, 15,
and 16, respectively. The curves are summarized together, without data
points, in Figure 17. The use of greased impact plates improved the
consistency of the determinations. A number of investigators have noted
and reported the tendency for particles to bounce or be reentrained from
impactor -stages. 1S~18 This tendency is accentuated in the higher velocity
stages intended to collect the finer particles. The net effect is to
make the aerosol appear to be finer than it actually is. The adhesive
coating oz the impact plates helps to reduce the bouncing or reentrain-
ment, but it is uncertain to what degree these phenomena were eliminated
in this investigation.
The distributions of Aerosol D (Figure 14) and Aerosol E (Figure 15)
are reasonably well represented as log-normal over the important parts of
the size ranges. The actual size distribution of Aerosol G (Figure 16)
was probably similar, but most (707o) of the material was finer than that
(i.e., finer than approximately 0.5 (J-m) collectable on the highest velocity
impactor stage. The portion of the distribution of Aerosol G lying below
0.8-1.0 p,m might, within the precision of measurement, be represented by
a straigh: line.
In all the particle-size distributions, the "spike" in the size
distribution of the coarsest l-27o of the aerosol evidently is fictitious.
The electron micrographs of the aerosol samples do not indicate that
these large particles actually existed. The indicated particle sizes
are the characteristic particle diameters for the impactor stages; that
is, the diameter of the particle that should be collected with 507»
efficiency. However, smaller particles are also collected although with
36
-------
LOW
MAGNIFICATION
0.2
0.4
1.0
HIGH
MAGNIFICATION
0.2 Mm
0.5
1.0
SA-4380-6O
FIGURE 13 ELECTRON MICROGRAPH OF AEROSOL G
-------
20
15
10
|
cc
LU
ui
o
h-
CC
<
Q.
1.0
0.1 L-
I T
O RUN 53 D (GREASED PLATES)
a RUN 57 D (GREASED PLATES)
A RUN 60 D (GREASED PLATES)
I
0.01 0.1 1 10 50 90 99
WEIGHT PERCENT FINER THAN INDICATED SIZE
99.9 99.99
SA-4380-52
FIGURE 14 PARTICLE-SIZE DISTRIBUTION OF AEROSOL D — MASS BASIS
(BY CASCADE IMPACTOR)
38
-------
20
15
10
E
a.
o:
01
O
tr
1.0
0.1
O RUN 62 D
O RUN 63 D
A RUN 73 D
(GREASED PLATES)
(GREASED PLATES)
(GREASED PLATES)
RUN OS-E-22 (DRY PLATES)
0.01 0.1 1 10 50 90 99
WEIGHT PERCENT FINER THAN INDICATED SIZE
99.9 99.99
SA-4380-53
FIGURE 15 PARTICLE-SIZE DISTRIBUTION OF AEROSOL E — MASS BASIS
(BY CASCADE IMPACTOR)
39
-------
20
10
E
LU
5
5
UJ
_l
y
DC
a.
1.0
0.1
O RUN 66 D (GREASED PLATES)
D RUN 68 D (GREASED PLATES)
A RUN 75 D (GREASED PLATES)
0.01 0.1
O
10 50 90 99
WEIGHT PERCENT FINER THAN INDICATED SIZE
99.9 99.99
SA-4380-54
FIGURE 16 PARTICLE-SIZE DISTRIBUTION OF AEROSOL G — MASS BASIS
(BY CASCADE IMPACTOR)
40
-------
10
_ I
E
a
< 1.0
Q
LU
_I
o
I-
oc
0.1
AEROSOL D
AEROSOL E
AEROSOL G
I I I I I
0.01 0.1 1 10 50 90
WEIGHT PERCENT FINER THAN INDICATED SIZE
99
I
99.9 99.99
SA-4380-55
FIGURE 17 SUMMARY CURVES — PARTICLE-SIZE DISTRIBUTIONS
OF AEROSOLS D, E, AND G (MASS BASIS)
41
-------
lower efficiencies. The total amounts of material collected on these
impactor stages were low, and could well be accounted for by collection
of small fractions of the coarser particles actually known to be present
in the aerosols.
I:E the curves in Figures 14 and 15 that represent the bulk of the
aerosols £.re extrapolated to the line indicating 99.997o of the total
aerosol, the intercepts correspond roughly to the sizes of the largest
particles actually observed in the electron micrograph fields (see
Table 3).
The variations in the aerosols produced by the generator are illus-
trated by the miniscrubber signature curves for Aerosol D shown in
Figure 18. As is noted in Appendix B, the aerosol did not show con-
tinuous variability over the band of variation. Instead, it appeared to
be reproduced in a number of discrete variations, each of which was pro-
duced with a certain frequency. Each of these variant forms was closely
replicated when it did appear, but the number of replications of a given
form varied. The variant form that appeared with the greatest frequency
was, of course, the "average" aerosol. In Figure 18 the middle curve
represents the "average" variant, and the other two represent the extreme
variants encountered.
In Figure 19, the signature curves for the average Aerosols D, E,
F, and G are presented together for comparison. The mode of variation of
Aerosol F differed substantially from that of the other three aerosols;
the variants were erratic and poorly replicated. This behavior was
evidently associated with the solution contamination discussed above. In
the case of Aerosol G, the variability was greater than that of Aerosols
D and E, probably because some contamination of the solution still existed.
ORIFICE SCRUBBER
The tests of the orifice scrubber with Aerosol D initially were
carried out in conjunction with studies of the reproducibility and sta-
bility of aerosol generation. The first 10 runs, OS-D-1 through OS-D-10,
were replf.cates (see Table 4) , and the replication was very satisfactory.
These 10 determinations are all represented by the single triangular
point in Figure 20. The variability of Aerosol D, as indicated by the
variation in the signatures, was greater than the variation in the per-
formance of the orifice scrubber. Later variations in the aerosol or in
the determination of orifice scrubber efficiency, or in both, were greater,
*
In all the tabulations of scrubber performance data (Tables 1-19) the
results were transcribed from computer printouts, which carried the
numbers t:o four significant digits. However, the presence of the four
digits is not to be interpreted as an indication of precision; the data
are actually valid to no more than three significant figures.
42
-------
CC
K
m
5
z
10.0
8.0
6.0
4.0
2.0
1.0
0.8
0.6
10
I I I I
20 40 60 80 100
PRESSURE DROP — cm WC
200
I
400
SA-4380-5
FIGURE 18 VARIATIONS IN AEROSQ.L D INDICATED BY MINISCRUBBER
SIGNATURE CURVES
43
-------
10
Z
D
1.0
CC.
tu
01
5
D
Z
0.1
10
AEROSOL D —
AEROSOL E _
AEROSOL F —
AEROSOL G
I I
100
PRESSURE DROP — cm WC
1000
SA-4380-15
FIGURE 19 MINISCRUBBER SIGNATURE CURVES FOR AEROSOLS D, E, F,
AND G
44
-------
TABLE It. PERFORMANCE OF ORIFICE SCRUBBER ON AEROSOL D
•P-
Ol
lest
code
OS-D
1
3
It
5
6
7
8
9
10
11
12
13
lit
15
16
17
'.18
19
20
21
22
23
24
25
26
27
28
36
37
38
39
40
41
42
43
44
45
46
47
49
50
51
.52
54
55
56
57
58
Contactor
orifice
size
foul
2.54
3.81
2.54
3.81
Gas flow
rate
(std m3/min)
2.370
• 2.370
2.368
2.368
2.368
2.385
2.385
2.385
2.385
2.340
2.349
1.912
2.384
1.604
1.800
1.347
.550
.214
.9997
.141
.449
.141
. 1.122
1.577
3.166
2.63
- 2.213
0.8635
0.8322
0..8836
1.447
1.347
1.248
2.945
2.699
2.223
2.058
2.637
2.637
3.004
2.703
1.850
1.541
2.900
2.630
1.757
1.639
1.938
Cas
velocity
in
orifice
(m/sec)
77.96
77.96
77.89
77.89
77.89
78.45
78.45
78.45
78.45
76.97
77.26
62.9
78.4
52.75
59.21
44.29
51.0
39.92
32.88
37.52
47.67
37.52
36.90
51.88
46.28
38.44
32.36
28.4
27.37
29.06
47.58
44.29
41.06
43.06
39.45
32.49
30.09
38.55
38.55
43.92
39.51
27.04
22.52
42.40
38.45
25.69
23.95
28.33
Liquid-to-
gas ratio
( liter /std m3)
1.55
1.55
1.57
1.57
1.57
1.56
1.56
1.56
1.56
1.617
3.867
1.386
5.558
20.39
8.957
11.24
2.441
3.119
3.786
45.39
2.611
3.318
0.337
20.73
2.391
2.879
3.42
74.96
77.77
73.25
10.47
10.68
12.13
2.57
2.805
3.406
3.677
5.741
5.741
10.77
. 11.97
17.50
21.00
11.16
12.30
18.42
19.75
1.953
Effective
friction
loss
(cm WC)
101.6
101.6
101.5
101.5
101.5
102.6
102.6
102.6
102.6
98.89
151.0
59.6
189.6
246.5
127.3
79.92
44.82
29.70
19.48
290.6
40.17
25.69
17.93
239.1
34.48
25.0
14.44
306.1
296.7
300.7
86.69
75.59
64.07
30.00
25.77
18.36
15.43
40.94
40.94
96.04
82.63
51.47
41.04
94.96
79.92
47.41
42.81
9.758
Contacting
power
(kWh/1000 m3)
2.768
2.768
2.765
2.765
2.765
2.795
2.795
2.795
2.795
2.694
4.114
1.624
5.166
6.716
3.469
2.177
1.231
0.8155
0.5350
7.915
1.103
0.7054
0.4924
6.513
0.9469
0.6865
0.3965
8.339
8.081
8.192
2.362 •
2.059
1.745
0.8171
0.7021
0.5001
0.4203
1.115
1.115
2.616
2.251
1.402
1.118
2.587
2.177
1.291
1.166
0.2658
Aerosol
generation
rate
(m£/min)
9.004
8.419
7.784
7.983
8.038
7.582
7.925
7.497
7.582
4.714
2.714
It. 226
2.243
2.315
3.313
4.934
5.747
5.524
5.65
3.092
5.436
4.651
5.135
1.966
5.086
4.985
5.392
4.006
3.785
4.037
4.685
4.556
4.82
5.229
5.231
5.21
5.121
5.851
15.56
9.854
11.98
24.76
14.47
3.473
4.803
5.651
5.751
5.409
Aerosol
concentration
(me/std m3)
In
3.799
3.552
3.287
3.371
3.394
3.179
3.323
3.144
3.179
1.788
1.155
2.21
0.9411
1.444
1.841
3.664
3.706
4.552
5.652
.2.71
3.75
4.077
4.577
1.246
1.606
1.896
2.436
4.64
4.548
4.568
3.239
3.384
3.861
1.775
1.938
2.344
2.488
2.219
5.898
3.280
4.433
13.39
9.392
1.198
1.826
3.216
3.510
2.791
Out
0.1267
0.1294
0.1140
0.1135
0.1166
0.1138
0.1138
0.1138
0.1152
0.07262
0.01702
0.2766
0.008652
0.007716
0.0383
0.1217
0.4389
0.9138
2.035
0.007705
0.4085
1.149
5.065
0.008675
0.356
0.623
1.309
0.01377
0.009952
0.009374
0.08766
0.06979
0.2071
0.4284
0.5654
0.9739
1.214
0.3004
0.722
0.9787
0.1626
0.6028
1.212
0.03972
0.07858
0.3589
0.4738
1.907
Collection
efficiency
(%)
96.66
96.36
96.53
96.63
96.57
96.42
96.58
96.38
96.38
95.94
98.53
87.48
99.08
99.47
97.92
96.68
88.16
79.93
63.98
99.72
89.11
71.83
--
99.30
77.84
67.14
46.28
99.70
99.78
99.79
97.29
97.94
94.64
75.88
70.83
58.45
51.19
86.45
87.75
97.02
96.33
95.50
87.10
96.68
95.70
88.84
86.50
31.70
Transfer
units
-------
10
t
z
tr
UJ
u.
g 1.0
<
DC
tL
UI
ffl
(J
0.1
I
I I I I I I I
I
10
100
EFFECTIVE FRICTION LOSS — cm WC
1000
SA-4380-10
FIGURE 20 ORIFICE SCRUBBER PERFOMANCE CURVE FOR AEROSOL D
46
-------
as indicated by the scatter of the data in Figure 20. However, these
variations were considered tolerable, and the performance curve was well
defined. The appearance of the two-branch performance curve is notable.
The same phenomenon had appeared under some conditions in the earlier
study with the same scrubber and a similar ammonium fluorescein aerosol.
However, the finding had been clouded by the instability of the aerosol
generator and the resulting problems of data interpretation. It would
obviously be possible to fit a single curved line to the data points of
Figure 20. However, past experience has shown scrubber performance
curves to be well represented by straight lines, and the data here are
certainly well represented by the two straight lines.
The energy consumption (or dissipation) was represented in these
tests by the "effective friction loss," expressed for convenience in
units of centimeters of water. The values used were actually equal to
the measured gas pressure drop across the scrubber with the equipment
under test. No appreciable part of the gas pressure drop resulted from
kinetic energy changes alone or from "ineffective friction loss" through
dry equipment. No significant energy was supplied from the liquid stream.
The same general behavior appeared during tests of Aerosol E (Figure
21 and Table 5). In all these tests the 0.17o ammonium fluorescein solu-
tion was prepared by diluting the stock solution with 0.3N ammonium
hydroxide solution, since the potential problems associated with the
practice had not been recognized. However, no substantial problems
appeared with Aerosol E. The single triangular data point in Figure 21
represents three replicate tests, OS-E-2, OS-E-4, and OS-E-5.
The performance tests of the orifice scrubber with Aerosol F (Figure
22 and Table 6) were completed before the sample analyses and data analysis
could be finished. Once the scatter in the data was noted and the SEM
observations demonstrated the nonspherical shape of the particles, the
studies described above revealed the contamination problem. The new
Aerosol G was then developed and tested.
In Figure 22, the performance curve for Aerosol F was drawn as a
single straight line, since the precision of the data justified nothing
more elaborate. It is possible that the actual curve should have been
two-branched as are the curves for the other three aerosols.
The performance curve for Aerosol G (Figure 23 and Table 7) took the
same characteristic form as those for the larger Aerosols D and E. During
the first four tests (see Table 7), a restriction was inadvertently left
in the water inlet, so that water was injected as a jet under substantial
upstream water pressures. The equivalent liquid-phase contacting power
was calculated, expressed in units of effective friction loss (cm WC),
and added to the effective friction loss corresponding to the gas pressure
drop. This procedure did improve the correlation of the data. The four
data points are shown as triangles in Figure 23. Not only was the pre-
cision of the data for Aerosol G much better than that for Aerosol F,
but the range of collection efficiencies (or transfer units) was strikingly
47
-------
10
t
DC
111
LL
1.0
QC
D
0.1
10
'O
100
EFFECTIVE FRICTION LOSS — cm WC
1000
SA-438O-11
FIGURH 21 ORIFICE SCRUBBER PERFORMANCE CURVE FOR AEROSOL E
48
-------
TABLE 5. PERFORMANCE OF ORIFICE SCRUBBER ON AEROSOL E
Test
code
OS-E
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
23
24
25
26
Contactor
orifice
size
(cm)
2.54
3.81
Gas flow
rate
(std m3/min)
2.263
2.384
2.405
2.379
2.370
2.309
2.309
1.991
1.087
1.145
0.9359
1.736
1.554
1.412
1.319
1.335
1.173
0.9378
0.9924
1.478
2.305
1.832
2.228
2.903
2.977
Gas
velocity
in
orifice
(m/sec)
74.45
78.42
79.10
78.25
77.96
75.93
75.93
65.48
35.74
37.65
30.78
57.09
51.13
46.44
43.37
43.90
38.57
30.85
32.64
21.61
33.70
26.78
32.58
42.44
43.52
Liquid-to-
gas ratio
(liter/std m3)
1.672
1.588
1.574
1.591
1.597
1.640
6.558
7.605
50.64
39.58
69.16
4.362
4.870
5.361
5.741
5.672
6.455
8.072
3.814
10.24
6.568
3.100
3.397
2.608
2.543
Effective
friction
loss
(cm WC)
92.65
103.9
101.6
103.6
103.6
92.79
188.0
136.1
291.0
240.8
324.8
74.23 -
57.84
46.06
43.10
43.66
35.34
25.86
19.70
14.57
30.39
10.17
17.07
21.46
30.60
Contacting
power
(kWh/1000 m3)
2.524
2.830
2.764
2.823
2.823
2.528
5.122
3.708
7.926
6.561
8.848
2.022
1.576
1.255
1.174
1.189
0.9627
0.7044
0.5365
0.3968
0.8277
0.2771
0.4649
0.5847
0.8336
Aerosol
generation
rate
(mH/min )
1.623
1.598
• 0.8966
1.498
1.486
1.747
1.186
1.270
1.287
1.090
1.218
1.423
1.537
1.625
1.520
0.8188
1.366
1.565
1.564
1.372
1.309
1.843
1.528
1.642
1.453
Aerosol
concentration
(mg/std m3)
In
0.7172
0.6701
0.3729
0.6298
0.6268
0.7567
0.5138
0.6380
1.185
0.9519
1.301
0.8201
0.9886
1.151
1.153
0.6135
1.164
1.669
1.576
0.9280
0.5678
1.006
0.6855
0.5655
0.4881
Out
0.07887
0.05972
0.03943
0.05645
0.05418
0 . 1044
0.01834
0.0417
0.01428
0.02209
0.008577
0.1580
0.2516
0.3648
0.4173
0.2369
0.5027
0.8692
0.9790
0.6959
0.2962
0.8091
0.5234
0.3518
0.2516
Collection
efficiency
(%)
89.00
91.09
89.42
91.04
91.36
86.20
96.43
93.46
98.79
97.68
99.34
80.73
74.55
68.31
63.80
61.39
56.83
47.93
37.88
25.01
47.84
19.58
23.65
37.80
48.45
Transfer
units
(Nt)
2.208
2.418
2.247
2.412
2.448
1.981
3.333
2.728
4.418
3.763
5.022
1.647
1.368
1.149
1.016
0.9516
0.840
0.6525
0.4762
0.2878
0.6509
0.2179
0.2698
0.4748
0.6626
VO
-------
10 ,
TT
i
1.0
tt
111
u.
u.
O
DC
m
m
0.1
0.03
O
O
I I I 11
10
1 I I I I I 1|
O O/
I I I I I I L
O
to
O
100
EFFECTIVE FRICTION LOSS — cm WC
1000
SA-4380-14
FIGURE 22 ORIFICE SCRUBBER PERFORMANCE CURVE FOR AEROSOL F
50
-------
TABLE 6. PERFORMANCE OF ORIFICE SCRUBBER ON AEROSOL F
Test
code
OS-F
1
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
22
23
24
25
Contactor
orifice
size
(cm)
2.54
3.81
2.54
Gas flow
rate
(std m3/min)
0.9359
0.8361
0.6681
0.6327
1.382
1.217
1.033
0.8775
0.6542
0.4359
2.763
2.396
2.379
1.499
2.122
1.850
1.850
1.894
1.894
0.7922
0.6750
0.9359
0.8254
Gas
velocity
in
orifice
(m/sec)
30.78
27.50
21.98
20.81
45.44
40.05
33.99
28.86
21.52
14.34
40.39
35.03
34.78
21.92
31.02
27.04
27.04
27.69
27.69
26.06
22.20
30.78
27.15
Liquid-to-
gas ratio
(liter/std m3)
69. 16
77.41
96.87
102.3
23.42
26.58
31.31
36.88
49.47
74.24
5.48
6.318
6.364
10. 10
7.135
8.185
6.139
3.996
1.998
81.70
95.89
69.16
78.42
Effective
friction
loss
(cm WC)
320.4
283.1
233.7
216.2
186.0
165.3
135.5
104.7
79.92
53.51
45.94
34.57
34.57
29.83
25.95
21.72
17.24
12.93
9.396
271.3
242.5
311.0
284.5
Contacting
power
(kWh/1000 m3)
8.727
7.712
6.365
5.889
5.066
4.502
3.690
2.852
2.117
1.458
1.252
0.9416
0.9416
0.8124
0.7068
0.5917
0.4696
0.3522
0.2559
7.391
6.605
8.472
7.749
Aerosol
generation
rate
(ms/min)
0.1407
0.1275
0.1305
0.1400
0.08994
0.09042
0.09715
0.1073
0.1109
0.1332
0.1213
0.1320
0.1290
0.1110
0.1749
0.1688
0.1611
0.1510
0.1471
0.1275
0.1279
0.1705
0.1412
Aerosol
concentration
(me/std m3)
In
0.1503
0.1525
0.1954
0.2213
0.0651
0.07427
0.0940
0.1223
0.1695
0.3055
0.04391
0.05509
0.05424
0.07403
0.08245
0.09126
0.08707
0.07972
0.07764
0.1610
0.1896
0.1822
0.1710
Out
0.004345
0.001155
0.0006657
0.01223
0.006563
0.009002
0.01749
0.03376
0.0557
0.09551
0.02515
0.03764
0.03697
0.06083
0.06459
0.07111
0.07291
0.07291
0.07414
0.009246
0.01135
0.004671
0.003451
Collection
efficiency
(%)
97. 11
99.24
99.66
94.47
89.92
87.88
81.39
72.40
67. 14
68.74
42.71
31.69
31.84
17.84
21.67
22.08
16.26
8.543
4.511
94.26
94.01
97.44
97.98
Transfer
units
(Nt)
3.544
4.883
5.682
2.896
2.295
2.110
1.681
1.287
1.113
1.163
0.5571
0.3811
0.3833
0.1965
0.2442
0.2495
0.1775
0.0893
0.04616
2.857
2.816
3.664
3.903
-------
c
u
K.
(E
UJ
03
0.01
10
100
EFFECTIVE FRICTION LOSS — cm WC
1000
SA-438O-16
FIGURtE 23 ORIFICE SCRUBBER PERFORMANCE CURVE FOR AEROSOL G
52
-------
TABLE 7. PERFORMANCE OF ORIFICE SCRUBBER ON AEROSOL G
Test
code
OS-G
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
Contactor
orifice
size
(cm)
2.54
3.81
2.54
Gas flow
rate
(std m3/min)
0.9359
1.442
1.217
1.004
1.721
1.350
0.8571
1.106
1.185
1.102
1. 106
2.245
1.883
0.9169
0.8875
1.004
1.366
1.050
1.646
1.269
1.145
1.025
1.012
1.286
Gas
velocity
in
orifice
(m/sec)
30.78
47.42
40.05
33.01
56.61
44.42
28.19
36.39
38.99
36.26
36.39
32.82
27.53
30.16
29.19
33.01
44.93
34.54
54.15
41.74
37.65
33.72
33.29
42.29
Liquid-to-
gas ratio
(liter/std m3)
69.16
26.94
31.90
38.70
8.797
11.21
17.66
6.842
3.193
3. 433
1.711
1.686
2.010
70.59
72.92
51.60
23.69
30.82
9.196
11.93
13.23
14.77
3.739
2.944
Effective
friction
loss
(cm WC)
317.8
359.1*
253.3
260.0*
201.8
208.5*
163.9
172.1*
119.2
81.28
39.96
33.88
25.77
21.55
18.62
12.59
9.31
321.7
298.0
243.8
188.6
133.2
110.3
69.76
59.33
49.91
19.22
29.74
Contacting
power
(kWh/1000 m3)
9.783
7.055
5.680
4.688
3.247
2.214
1.089
0.9228
0.7021
0.5870
0.5072
0.3428
0.2536
8.764
8.118
6.642
5.136
3.627
3.004
1.900
1.616
1.360
0.5236
0.8101
Aerosol
generation
rate
(mg/min)
0.1076
0.06928
0.06299
0.06782
0.07753
0.08295
0.1121
0.1080
0.1023
0.1009
0.09813
0.1038
0.1013
0.08095
0.08072
0.06721
0.04397
0.05430
0.05838
0.06730
0.06838
0.07013
0.07248
0.07473
Aerosol
concentration
(mg/std m3)
In
0.1150
0.04806
0.05174
0.06758
0.04505
0.06142
0.1308
0.09759
0.08629
0.09156
0.08869
0.04623
0.05376
0.08829
0.09094
0.06697
0.03218
0.0517
0.03546
0.05303
0.05974
0.06842
0.07161
0.05811
Out
0.009631
0.009007
0.01270
0.01823
0.02759
0.02572
0.08709
0.06837
0.07102
0.07839
0.08416
0.04444
0.05352
0.01108
0.01194
0.01220
0.009470
0.01837
0.01397
0.02667
0.0327
0.04050
0.06426
0.04574
Collection
efficiency
(%)
91.63
81.26
75.46
73.03
38.76
58.12
33.40
29.94
17.70
14.38
5.11
3.867
0.449
87.45
86.87
81.79
70.58
64.47
60.06
49. 71
45.27
40.81
10.27
21.29
Transfer
units
-------
different from that for Aerosol F. In fact, the nominal particle sizes
of the two aerosols should have been essentially the same. The two
solutions contained the same concentration of ammonium fluorescein, and
presumably the sizes of the droplets produced in the nebulizer should
have been the same or nearly so. The presence of the silica in Aerosol F
should not have significantly changed the particle size, although it
certainly did affect the shape.
The orifice scrubber performance curves for the four aerosols are
summarized together without data points in Figure 24. They may be compared
with the corresponding aerosol signature curves of Figure 19, which quali-
tatively show the same general trends. The deviations between the signa-
ture curves for the different aerosols are relatively greater than those
between the corresponding orifice scrubber performance curves, demonstrat-
ing the greater sensitivity of the miniscrubber to changes in the aerosol.
MULTIPLE-ORIFICE SERIES SCRUBBER
Once the performance of the orifice scrubber was established as a
baseline, testing of the multiple-orifice series contactor was undertaken.
The cata for the tests on Aerosols D, E, and G are given in Tables 8, 9,
and 10, respectively. These data have been plotted in Figures 25, 26, and
27; in each figure the corresponding baseline performance curve for the
orifice scrubber is shown for reference.
In the tests on Aerosol D (Figure 25), the data for the multiple-
orifice series scrubber fitted the upper end of the orifice scrubber
performance curve as well as did the original data for the orifice
scrubber itself. However, along the lower part of the performance curve,
the multiple-orifice scrubber data generally tended to fall below the
curve. The significance of this behavior was not obvious from Figure 25
alone, However, the same trend appeared in accentuated degree with
Aerosol 3 (see Figure 26), and was very marked with Aerosol G (Figure 27).
With Aerosol G, the performance of the multiple-orifice contactor appeared
to attain that of the orifice scrubber only at the lowest and highest ends
of the performance range. In Figure 27, a single straight line has been
used to represent approximately the performance of the multiple-orifice
scrubber; but it appears that, except for the two highest points, the
data points would be better fitted by another two-branched curve. From
the analogy to the performance curves for Aerosols D and E, one might
reasonably expect the performance curves in Figure 27 to merge at the
upper ends, but this conjecture could not be confirmed because the limi-
tations of the equipment prevented operation at higher levels of effective
friction loss.
SPRAY AND SPRAY-ORIFICE SCRUBBERS
The test data for the spray and spray-orifice scrubbers are presented
in Table,-? 11-19, and in Figures 28-36. Because all the scrubber arrange-
ments used both gas-phase and liquid-phase contacting power, the number of
54
-------
10
EFFECTIVE FRICTION LOSS — cm WC
FIGURE 24 SUMMARY OF ORIFICE SCRUBBER PERFORMANCE CURVES
FOR AEROSOLS D, E, F, AND G
55
-------
TABLE 8. PERFORMANCE OF MULTIPLE-ORIFICE SERIES SCRUBBER ON AEROSOL D
Test
code
MOS-D
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
Contactor
orifice
size
(cm)
3.36
equivalent
Gas flow
rate
(std m3/min)
0.8361
0.6502
.0.5497
0.2651
1.133
0.9776
1.716
1.412
1.025
0.9264
0.9121
1.217
0.9771
1. 114
1.046
1.182
1.137
1.067
0.9816
0.9544
1. 217
1.106
Gas
velocity
in
ori f ice
(ml sec)
15.71
12.22
10.33
4.981
21.29
18.37
32.25
26.53
19.26
17.41
17. 14
22.88
18.36
20.94
19.66
22.21
21.37
20.04
18.45
17.94
22.88
20.79
Liquid- to-
gas ratio
(liter/std m3)
77.41
99.54
117.7
244.2
28.56
33. 10
7.543
10.72
14.77
8. 171
8.300
3.109
3.874
3.397
14.47
6.406
6.657
7.452
8.098
1.586
6.218
6.842
Effective
friction
loss
(cm WC)
330.5
281. 1
240.3
188.4
157. 1
133.0
107. 7
83.71
53.51
27.09
25.86
34.48
20.43
30.17
50.17
43.01
39.65
34.61
30.43
20.04
44.82
36.98
Contact ing
power
(kWh/1000 m3)
9.003
7.657
6.546
5. 133
4.280
3.624
2.933
2.280
1.458
0.7380
0.7044
0.9392
0.5565
0.8218
1.367
1.172
1.080
0.9428
0.8289
0.5459
1.221
1.007
Aerosol
generation
rate
(mg/min)
16.31
5.603
6.094
4.231
5.391
4.897
5.046
5.241
5.000
5.132
8.645
5.207
5.413
5.156
5.397
5.267
5.458
5.539
5.409
5.242
5.183
5.179
Ae rosol
concentration
(ma/std m3)
In
19.51
8.617
11.09
15.96
4.757
5.009
2.940
3.712
4.878
5.540
9.479
4.277
5.540
4.628
5.160
4.457
4.800
5.194
5.511
5.492
4.257
4.681
Out
0.01406
0.02392
0.04428
0.1245
0.05589
0.0939
0.05532
0.1475
0.5089
1.691
1.726
1.241
2.690
1.560
0.5227
0.6899
0.8511
1.121
1.398
3.243
0.6667
0.9481
Collection
ef f ic iency
(%)
99.93
99.72
99.60
99. 22
98.83
98. 13
98. 12
96.03
89.57
76.61
81.79
70.98
51.45
66.29
89.87
84.52
82.27
78.41
74. 63
40.96
84.34
79.75
Transfer
units
(Nt)
7.235
5.887
5.523
4.853
4.444
3.977
3.973
3.225
2.260
1.453
1.703
1.237
0.7226
1.088
2.290
1.866
1.730
1.533
1.371
0.5270
1.854
1.597
Ui
01
-------
TABLE 9. PERFORMANCE OF MULTIPLE-ORIFICE SERIES SCRUBBER ON AEROSOL E
Test
code
MOS-E
1
2
3
. 4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Contactor
orifice
size
(cnO
3.36
equivalent
Gas flow
rate
(std ml/min)
0.8307
0.7393
1.210
1.067
1.182
1.319
1.217
0.9359
0.7207
1.382
0.7808
0.6542
0.9359
0.7692
0.7514
1. 163
1.046
0.9452
1.004
0.9590
Gas
velocity
in
orifice
(m/sec)
15.61
13.89
22.75
20.04
22.21
24.78
22.88
17.59
13.54
25.96
14.67
12.29
17.59
14.46
14.12
21.86
19.66
17.76
18.86
18.02
Liquid-to-
gas ratio
(liter/std m3)
77.91
87.54
37.43
42.48
32.86
24.54
26.58
34.58
44.90
10.96
19.39
23.14
4.044
84.14
86.13
13.01
•14.47
16.02
7.543
3.947
Effective
friction
loss
(cm WC)
335.9
312.4
277.7
241. 1
215.4
188.3
161.2
134.1
105.0
78.57
34.05
25.86
20.34
320.6
311.6
65.70
51.72
43.06
30.17
21.64
Contacting
power
(kWh/1000 m3)
9.151
8.509
7.564
6.568
5.867
5.129
4.391
3.653
2.860
2.140
0.9275
0.7044
0.5541
8.734
8.487
1.790
1.409
1.173
0.8218
0.5894
Aerosol
generation
rate
(mR/min)
1.924
1.777
1.690
1.652
1.658
1.633
1.672
1.645
1.735
1.706
1.647
1.673
1.709
1.371
1.569
1.485
1.506
1.464
1.444
1.463
Aerosol
concentra tion
(mR/std m3)
In
2.316
2.404
1.396
1.548
1.403
1.238
1.374
1.758
2.407
1.235
2.110
2.558
1.826
1.782
2.088
1.276
1.440
1.549
1.439
1.525
Out
0.004013
0.009310
0.01949
0.03072
0.04458
0.04057
0.06667
0.1189
0.2423
0.2423
1.093
1.550
1.331
0.006806
0.006388
0.3737
0.5509
0.7536
0.8170
1.099
Collection
efficiency
(%)
99.83
99.61
98.60
98.02
96.82
96.72
95.15
93.24
89.94
80.38
48.19
39.41
27.10
99.62
99.69
70.72
61.74
51.35
43.23
27.96
Transfer
units
(V
6.358
5.554
4.272
3.920
3.449
3.419
3.026
2.694
2.296
1.629
0.6575
0.5010
0.3161
5.568
5.789
1.228
0.9607
0.7205
0.5661
0.3279
Cn
-------
TABLE 10. PERFORMANCE OF MULTIPLE-ORIFICE SERIES SCRUBBER ON AEROSOL G
Test
code
MOS-G
1
2
3
4
5
6
8
9
10
11
12
13
14
15
16
Contactor
orifice
size
(cm)
3.36
equivalent
Ga s £ low
rate
(std m3/min)
1.025
0.7015
1.054
1.442
1.319
1.145
1.838
1.634
1.252
1.067
1.269
1.122
0.9948
0.9816
0.9359
Gas
velocity
in
orifice
(m/sec)
19.26
13.18
19.81
27.09
24.78
21.51
34/55
30.70
23.53
20.04
23.85
21.08
18.69
18.45
17.59
Liquid-to-
gas ratio
(liter/std m3)
63.14
92.26
42.97
22.45
24.54
28.27
8.235
9.268
12.09
14.20
5.965
6.747
7.609
3.856
2.022
Effective
friction
loss
(cm WC)
324.3
297.3
243.6
214.0
189.6
162.6
132.1
106.5
67.32
55.54
49.56
38.79
30.17
24.14
19.96
Contacting
power
(kWh/1000 m3)
8.834
8.099
6.634
5.830
5.166
4.428
3.598
2.900
1.834
1.513
1.350
1.057
0.8218
0. 6575
0.5436
Aerosol
genera tion
rate
(tnR/min)
0.1589
0.1081
0.09769
0.08944
0.08766
0.08306
0.0804
0.08116
0.08824
0.08342
0.08025
0.07837
0.07846
0.07429
0.07550
Aerosol
concent ra tion
(mg/std iiH)
In
0. 1550
0.1541
0.09266
0.06204
0.06648
0.07255
0.04373
0.04968
0.07047
0.07821
0.0'6324
0.06985
0.07886
0.07568
0.08068
Out
0.01460
0.01677
0.02202 "
0.01664
0.02267
0.02851
0.01809
0.02397
0.04553
0.05397
0.04652
0.05532
0.06532
0.06730
0.07340
Collection
efficiency
(%)
90.58
89.12
76.24
73.18
65.91
60.70
58.65
51.75
35.39
31.00
26.43
20.81
17.18
11.07
9.01
Transfer
units
(Nt)
2.362
2.218
1.437
1.316
1.076
0.9341
0.8831
0.7288
0.4368
0.3710
0.3069
0.2333
0.1885
0.1173
0.0945
CO
-------
101 I I I I I I I I I I I I I I I I J_
oc
111
m
0.1
O
x
x"
t
z
c:
S,oL .' O
I //0
ORIFICE
SCRUBBER
o
10 100 1000
EFFECTIVE FRICTION LOSS — cm WC
SA-4380-17
FIGURE 25 MULTIPLE-ORIFICE SERIES SCRUBBER PERFORMANCE ON
AEROSOL D
59
-------
101 I I I I I I I I I I I I I I I 14-1
2
D
cc
S 1.0
c
00
5
3
pf*5
*T
\;<5''
ORIFICE Qx
X
SCRUBBER
/0°
/
— /
/
Q ^ | | 1 I I I I I I
10 100 1000
EFFECTIVE FRICTION LOSS — cm WC
SA-4380-18
FIGURE 26 MULTIPLE-ORIFICE SERIES SCRUBBER PERFORMANCE ON
AEROSOL E
60
-------
CO
i
cc
S i.o
CO
< 0.8
GC
0.6
0.4
0.2
0.1
1 I I I I
10
ORIFICE *' *
SCRUBBER S
-' 'o
V
100
EFFECTIVE FRICTION LOSS — cm WC
1000
SA-4380-20
FIGURE 27 MULTIPLE-ORIFICE SERIES SCRUBBER PERFORMANCE
ON AEROSOL G
61
-------
TABLE 11. PERFORMANCE OF CONFIGURATION SSI SPRAY SCRUBBER ON AEROSOL D
Test
code
SS1-D
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Contactor
line
size
(cm)
5.842
Gas flow
rate
(std m3/min)
4.571
4.172
3.833
4.383
2.430
4.597
2.011
2.072
3.922
3.214
1.894
1.217
1.025
1.145
1.436
1.980
Gas
velocity
in
line
Cm/sec)
28.42
25.94
23.83
27.25
15.11
28.58
12.51
12.89
24.38
19.99
11.78
7.57
6.373
7.118
8.927
12.31
Water rate
(liter /rain)
8.516
8.327
8.138
5.299
3.785
9.955
9.652
8.478
39.48
39.48
39.48
39.48
39.48
35.92
27.18
15.14
Water
pressure
(atm)
5.511
5.477
5.443
2.347
0.8505
13.74
13.61
10.21
10.07
10.07
10.07
10.07
10.07
8.505
4.763
1.225
Liquid-to-
gas ratio
(liter/std m3)
1.863
1.996
2.123
1.209
1.557
2.165
4.798
4.091
10.07
12.28
20.84
32.43
38.51
31.38
18.93
7.646
Effective
friction
loss
(cm WC)
26.55
22.24
19.01
16.46
3.62
27.50
7.456
6.724
72.20
54.32
30.00
21.55
19.57
18.19
14.91
8.017
Contacting power
(kWh/1000 m3)
Gas
PG
0.7232
0.6058
0.5178
0.4485
0.09862
0.7491
0.2031
0.1832
1.967
1.480
0.8171
0.5870
0.5330
0 .4954
0 .4062
0.2184
Liquid
PL
0.2891
0.3078
0.3254
0.07991
0.03729
0.8379
1.838
1.175
2.854
3.482
5.908
9.192
10.92
7.513
2.539
0.2636
Total
PT
1.012
0.9136
0.8431
0.5284
0.1359
1.587
2.041
1.359
4.821
4.961
6.725
9.779
11.45
8.009
2.945
0.4820
"/.
PL
28.6
33.7
38.6
15.1
27 .4
52.8
90.1
86.5
59.2
70.2
87.9
94.0
95.4
93.8
86.2
54.7
Aerosol
generation
rate
(mg/min)
5.946
5.601
5.685
5.434
5.559
5.983
5.645
5.713
5.778
6.149
6.100
6.036
6.063
5.825
5.625
5.783
Aerosol
concentration
(mfi/std tn )
In
1.301
1.342
1.483
1.240
2.287
1.301
2.806
2.757
1.473
1.913
3.220
4.958
5.915
5.088
3.918
2.920
Out
0.3075
0.3660
0.4791
0.4548
1.957
0.2287
1.254
1.385
0.05078
0.1052
0.2962
0.4530
0.4661
0.6272
1.092
1.840
Collection
efficiency
«)
76.36
72.74
67.70
63.33
14.42
82.42
55.31
49.76
96.55
94.50
90.80
90.86
92.12
87.67
72.12
37.00
Transfer
units
(V
1.442
1.300
1.130
1.003
0.1558
1.739
0.8054
0.6883
3.368
2.900
2.386
2.393
2.541
2.093
1.277
0.4621
TABLE 12. PERFORMANCE OF CONFIGURATION SS2 SPRAY SCRUBBER ON AEROSOL D
Test
code
SS2-D
1
2
3
4
5
6
7
8
9
Contactor
line
size
(cm)
5.842
Gas flow
rate
(std m /tnin)
1.350
1.757
1.350
1.757
1.286
1.861
1.910
1.959
2.291
Gas
velocity
in
line
(m/sec)
8.397
10.93
8.397
10.93
7.995
11.57
11.87
12.18
14.24
Water rate
(Hter/min)
22.67
15.22
22.97
15.22
24.26
12.94
11.96
11.54
3.785
Water
pressure
(atm)
12.25
5.171
12.11
5.171
13.61
3.946
2.926
2.705
0.1021
Liquid-to-
gas ratio
(liter/std m3)
16.79
8.659
17.01
8.659
18.87
6.956
6.263
5.893
1.652
Effective
friction
loss
(era WC)
16.81
11.21
16.81
11.21
17.24
9.827
9.094
8.404
3.965
Contacting power
(kWh/1000 m3)
Gas
JG
0.4579
0.3053
0.4579
0.3053
0.4696
0.2677
0.2477
0.2289
0.1080
Liquid
p
PL
5.789
1.261
5.801
1.261
7.229
0.7728
0.5159
0.4487
0.00475
Total
PT
6.246
1.566
6.259
1.566
7.698
1.040
0.7636
0.6776
0.1128
%
PL
92.7
80.5
92.7
80.5
93.9
74.3
67.6
66.2
4.2
Aerosol
generation
rate
(mE/min)
5.689
5.874
5.719
5.525
5.531
4.939
5.395
5.515
9.341
Aerosol
concentration
(tng/std nP)
In
4.212
3.343
4.235
3.144
4.301
2.654
2.825
2.815
4.078
Out
0.5121
1.092
0.5200
1.092
0.3989
1.127
1.283
1.369
2.028
Collection
efficiency
(7.)
87.84
67.33
87.72
65.26
90.73
57.54
54.59
51.35
50.27
Transfer
units
fN 1
(V
2.107
1.119
2.097
1.057
2.378
0.8567
0.7895
0.7206
0.6985
-------
TABLE 13. PERFORMANCE OF CONFIGURATION SS3 SPRAY SCRUBBER ON AEROSOL D
Test
code
SS3-D
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Contactor
line
size
(cm)
5.842
Gas flow
rate
(std m3/min)
2.413
2.437
2.305
2.236
2.170
2.300
2.447
4.676
4.334
3.973
3.563
2.900
4.693
4.292
3.882
3.333
2.879
2.282
Gas
velocity
in
line
(m/sec)
15.01
15.15
14.33
13.90
13.49
14.30
15.22
29.07
26.95
24.70
22.15
18.03
29.18
26.69
24.14
20.72
17.90
14.19
Water rate
(Hter/min)
19.42
17.98
16.50
13.93
11.35
3.785
21.04
20.70
20.70
20.70
20.70
20.70
15.22
15.22
15.22
15.22
15.22
15.22
Water
pressure
(atm)
11.98
9.798
7.485
5.273
2.211
0.0510
13.61
13.61
13.61
13.61
13.61
13.61
6.804
6.804
6.804
6.804
6.804
6.804
Liquid-to-
gas ratio
(liter/std m3)
8.046
7.377
7.160
6.230
5.233
1.646
8.599
4.428
4.777
5.211
5.811
7.139
3.242
3.545
3.920
4.565
5.285
6.669
Effective
friction
loss
(cm WC)
2.370
3.060
3.879
4.741
5.344
3.879
1.638
24.14
20.17
15.86
11.03
5.258
25.86
21.55
17.24
12.07
8.62
4.396
Contacting power
(kWh/1000 m3)
Gas
PG
0.06457
0.08336
0.1057
0.1291
0.1456
0.1057
0.04461
0.6575
0.5495
0.4320
0.3006
0.1432
0.7044
0.5870
0.4696
0.3287
0 . 2348
0.1198
Liquid
PL
2.712
2.035
1.509
0.9248
0.3257
0.002365
3.294
1.696
1.830
1.996
2.226
2.735
0.6211
0.6791
0.7508
0.8745
1.012
1.277
Total
PT
2.777
2.118
1.614
1.054
0.4713
0.108
3.339
2.354
2.379
2.428
2.527
2.878
1.325
1.266
1.220
1.203
1.247
1.397
%
PL
97.7
96.1
93.5
87.7
69.1
2.2
98.7
72.0
76.9
82.2
88.1
95.0
46.9
53.6
61.5
72.7
81.2
91.4
Aerosol
generation
rate
(rag/rain)
5.254
5.657
5.399
5.292
5.498
5.476
5.315
5.623
5.449
5.706
5.568
5.610
5.527
5.805
5.744
5.921
5.639
5.659
Aerosol
concentration
(mg/std m3)
In
2.177
2.321
2.343
2.367
2.533
2.381
2.172
1.203
1.257
1.436
1.563
1.934
.178
.353
.480
.777
.959
2.480
Out
0.5546
0.7887
1.031
1.378
1.716
0.7245
0.4182
0.1881
0.2440
0.3135
0.3790
0.4511
0.2235
0.3135
0.4287
0.5722
0.8494
1.228
Collection
efficiency
(%)
74.53
66.02
55.97
41.78
32.27
69.58
80.75
84.36
80.60
78.17
75.74
76.68
81.02
76.82
71.03
67.79
56.63
50.48
Transfer
units
-------
TABLE 14. PERFORMANCE OF CONFIGURATION SS4 SPRAY SCRUBBER ON AEROSOL D
Test
^«^P
334-iJ
i
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
*30
*31
*32
*33
*34
*35
36
37
38
39
40
41
42
43
44
45
46
47
48
Contactor
line
size
(r~)
^ fOt. t
Gas f low
rate
J\tr" -3/T-in)
'».417
3.790
3.045
2.578
2.217
1.792
4.557
4.006
3.595
3.157
2.562
1.984
1.286
1.485
1.286
1.286
1.286
1.286
3.540
3.540
3.540
3.540
3.540
2.610
2.610
2.610
2.610
2.610
2.610
0.7207
0.7207
0.7207
0.7207
0.7207
1.872
2.763
4.074
4.650
4.641
4.074
2.763
1.872
0.7207
0.7207
0.7207
0.7207
0.7207
1.872
Gas
velocity
in
line
|V,/«Prl
27."
23.56
18.93
16.03
13.79
11.15
28.34
24.91
22.35
19.63
15.93
12.34
7.995
9.233
7.995
7.995
7.995
7.995
22.01
22.01
22.01
22.01
22.01
16.23
16.23
16.23
16.23
16.23
16.23
4.481
4.481
4.481
4.481
4.481
11.64
17.18
25.33
28.91
28.86
25.33
17.18
11.64
4.481
4.481
4.481
4.481
4.481
11.64
diter/r"inl
17 7Q
17.49
17.34
17.34
17.34
17.34
13.25
13.25
13.25
13.25
13.25
13.25
17.15
15.06
15.06
12.68
9.463
7.570
17.34-
15.22
12.60
9.463
7.570
18.36
15.22
12.60
11.13
9.463
7.570
17.34
15.22
12.68
9.463
7.570
7.570
7.191
7.570
7.381
9.463
9.463
9.463
9.463
17.34
15.22
12.76
9.463
7.570
7.267
(atm)
13.
85.89
79.99
71.12
64.02
58.60
55.13
81.24
73.58
67.51
61.72
52.12
46.55
48.75
43.87
40.70
30.80
17.67
11.12
75.68
70.86
64.74
61.60
55.35
63.23
56.36
49.49
45.53
39.83
33.6
65^95
58.83
43.62
28.53
14.65
27.86
36.78
65.20
75.76
77.75
69.45
40.96
25.90
61.74
49.54
36.98
19.06
6.998
18.07
Transfer
units
(Nt)
1.958
1.609
1.242
1.022
0.8818
0.8014
1.674
1.331
1.124
0.9602
0.7364
0.6264
0.6685
0.5775
0.5225
0.3682
0 . 1944
0.1179
1.414
1.233
1.042
0.9570
0.8064
1.000
0.8293
0.6831
0.6075
0 . 5080
0.4096
1.077
0.8875
0.5730
0.3359
0.1585
0.3266 '
0.4586
1.056
1.417
1.503
1.186
0.5270
0.2997
0.9608
0.6839
0.4617
0.2114
0.0726
0.1994
05
Sweep air 6 liters/min.
-------
TABLE 15. PERFORMANCE OF CONFIGURATION SS4 SPRAY SCRUBBER ON AEROSOL G
Test
code
SS4-G
1
2
3
4
5
6
7
8
9
10
11
12
13
lit
15
16
17
18
19
20
Contactor
line
size
(cm)
5.842
Gas flow
rate
(std tn3/min)
0.7393
1.286
1.872
2.907
4.684
4.654
2.907
1.872
1.286
0.7207
0.7207
1.286
1.872
2.907
4.571
4.490
2.907
1.872
1.286
0.7207
Gas
velocity
in
line
(m/sec)
4.597
7.995
11.64
18.08
29.13
28.94
18.08
11.64
7.995
4.481
4.481
7.995
11.64
18.08
28.42
27.92
18.08
11.64
7.995
4.481
Water rate
(liter/min)
7.570
7.570
7.381
7.381
7.267
9.463
9.463
9.652
9.841
9.652
12.76
12.68
12.76
12.68
12.76
17.34
17.34
17.34
17.34
17.34
(atm)
1.769
1.837
1.769
1.701
1.701
3.334
3.402
3.402
3.402
3.334
6.804
6.804
6.804
6.804
6.804
13.61
13.61
13.61
13.61
13.61
gas ratio
(llter/std m3)
10.24
5.887
3.942
2.539
1.551
2.033
3.255
5.155
7.653
13.39
17.70
9.861
6.813
4.361
2.791
3.861
5.963
9.26
13.48
24.05
Effective
loss
(cm WC)
1.638
2.931
4.569
10.09
25.00
28.45
11.98
5.689
3.189
1.948
2.672
4.569
7.241
14.05
32.41
38.45
18.79
10.60
7.499
4.051
Contacting power
(kWh/1000 m3)
Gas
PG
0.04461
0.07983
0.1244
0.2747
0.6809
0.7749
0.3264
0.1550
0.08688
0.05307
0.07279
0. 1244
0. 1972
0.3827
0.8829
1.047
0.5119
0.2888
0.2043
0.1104
Liquid
PL
0.5099
0.3045
0.1963
0.1216
0.07429
0. 1908
0.3117
0.4938
0.7330
1.257
3.390
1.889
1.305
0.8354
0.5346
1.479
2.284
3.547
5.165
9.215
Total
PT
0.5546
0.3843
0.3208
0.3963
0.7552
0.9657
0.6381
0.6487
0.8199
1.310
3.463
2.013
1.502
1.218
1.417
2.526
2.796
3.836
5.369
9.325
'/.
PL
92.0
79.2
61.2
30.7
9.8
19.8
48.9
76. 1
89.4
96.0
97.9
93.8
86.9
68.6
37.7
58.6
81.7
92.5
96.2
98.8
rate
(me/min)
0.1171
0.1191
0.1770
0.1141
0.1223
0.1184
0.1084
0.1007
0.09582
0.09387
0.09395
0.09017
0.09151
0.09053
0.1079
0.09663
0.1496
0.1463
0.1101
0. 1048
(ms/s
In
0.1583
0.09263
0.09452
0.03923
0.02611
0.02544
0.03729
0.05381
0.07452
0.1302
0.1304
0.07012
O.OA888
0.03114
0.02361
0.02152
0.05147
0.07815
0.08562
0.1455
A m3)
Out
0. 1445
0.09014
0.06702
0.03886
0.02289
0.02154
0.03539
0.05309
0.07402
0.1225
0.1156
0.06702
0.04646
0.02914
0.02011
0.01604
0.04300
0.06755
0.07402
0. 1158
efficiency
(»
8.74
2.70
29.1
0.956
12.34
15.34
5.09
1.34
0.67
5.93
1 11.33
4.43
4.95
6.41
14.85
25.47
16.46
13.57
13.54
20.39
units
(Nt)
0.0914
0.0273
0.3439
0.0096
0.1317
0.1665
0.0522
0.0135
0.0067
0.0612
0.1202
0.0453
0.0507
0.0662
0.1608
0.2940
0.1799
0.1458
0.1455
0.2280
o\
-------
TABLE 16. PERFORMANCE OF CONFIGURATION SS5 SPRAY SCRUBBER ON AEROSOL D
Test
code
SS5-D
1
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
15
26
27
28
29
30
31
32
33
Contactor
line
size
fcml
5.842
Gas flow
rate
(std m3/min}
1.905
4.606
4.259
3.642
3.176
2.710
1.634
1.350
0.9816
1.442
1.302
0.8875
1.217
0.9816
0.4121
2.180
1.527
0.9816
4.268
3.901
4.544
3.973
3.163
2.562
4.579
3.848
3.004
2.314
2.309
0.9590
0.2901
Gas
velocity
in
1 ine
(m/ sec}
11.85
28.64
26.48
22.65
19.75
16.85
10.16
8.397
6. 103
8.964
8.097
5.519
7.570
6.103
2.562
13.55
9.495
6. 103
26.54
24.25
28.25
24.70
19.67
15.93
28.47
23.93
18.68
14.39
14.35
5.963
1.804
Water rate
(liter/mini
21.80
21.88
21.88
21.88
21.88
21.88
21.88
21.88
21.88
15.52
15.52
15.52
13.63
13.25
13.63
18.81
18.81
18.81
21.80
21.80
15.86
15.86
15.86
15.86
10.67
11.13
10.94
11.05
11. 13
11. 13
11.05
Water
pressure
(atml
13.61
13.61
13.61
13.61
13.61
13.61
13.61
13,61
13.61
6.804
6.804
6.804
5.103
5.103
5.103
10.21
10.21
10.21
13.61
13.61
6.804
6.804
6.804
6.804
3.266
3.402
3.402
3.402
3.402
3.402
3.402
Liquid-to-
gas ratio
(liter/std m3')
11.44
4.750
5. 137
6.007
6.888
8.072
13.39
16.20
22.29
10.76
11.92
17.48
• 11.19
13.50
33.07
8.631
12.32
19.16
5.108
5.589
3.490
3.992
5.014
6.189
2.331
2.892
3.641
4.776
4.820
11.60
38.10
Effective
friction
loss
fern WC1
-8.810
13.53
9.913
4. 224
0.3448
3.620
-10.30
-12.59
-15.00
-3.776
-4. 224
-5.818
-2.758
-3.707
-5.301
-5.758
-7.068
-9.999
10.34
6.896
15.52
10.52
5.172
1.810
15.43
10.34
5.215
2.457
2.414
-1.784
-2.974
Contacting power
(kWh/1000 m3)
Gas
pr,
-0.24
0.3686
0.270
0.1151
0.00939
0.0986
-0.2806
-0.3428
-0.4086
-0.1028
-0.1151
-0.1585
-0.0751
-0.1010
-0.1444
-0.1569
-0.1925
-0.2724
0.2818
0.1878
0.4227
0.2865
0 . 1409
0.0493
0.4203
0.2818
0.1421
0.0669
0.0657
-0.0486
-0.0810
Liquid
•Y
4.384
1.820
1.968
2.301
2.639
3.092
5.131
6.206
8.539
2.062
2.283
3.349
1.608
1.939
4.751
2.480
3.540
5.507
1.957
2.141
0.6685
0.7646
0.9604
1.186
0.2143
0.2769
0.3487
0.4575
0.4617
1. Ill
3.649
Total
PT
4.144
2.188
2.238
2.416
2.648
3.191
4.850
5.863
8.130
1.959
2.168
3.191
1.533
1.838
4.606
2.323
3.347
5.234
2.239
2.329
1.091
1.051
1. 101
1.235
0.6346
0.5587
0.4908
0.5244
0.5274
1.063
3.568
%
•Y
105.8
83.15
87.93
95.24
99.65
96.91
105.8
105.8
105.0
105.2
105.3
105.0
104.9
105.5
103.1
106.8
105.8
105.2
87.41
91.93
61.27
72.75
87.21
96.01
33.77
49.57
71.06
87.24
87. 53
104.6
102. 3
Aerosol
genera tion
rate
(ma/mini
3.323
4.158
3.944
3.702
3.686
3.588
3.859
3.198
3.052
3.953
3.114
3.045
3.922
3.232
2.748
4.126
3.877
2.898
3.192
3.176
3.252
3.190
3.171
3.014
3.739
1.128
4.081
2.764
2.964
3.010
2.629
Aerosol
concentration
(mR/std m3)
In
1.744
0.9027
0.9260
1.016
I. 161
1.324
2.362
2.368
3. 109
2.742
2.391
3.431
3. 222
3.293
6.669
1.893
2.539
2.953
0.7478
0.8143
0.7157
0.8029
1.002
1.176
0.8165
0.2931
1.358
1.194
1.284
3. 138
9.064
Out
0.1355
0. 1668
0.1813
0.2021
0.2118
0.1774
0. 1178
0.0967
0.0949
0.6102
0.5178
0.4717
0.9711
0.7861
0.7398
0.2681
0. 2413
0. 1831
0. 1554
0.1693
0.2193
0.3051
0.4253
0.4623
0.3145
0.1527
0.8322
1. 104
0.8651
1.417
2. 186
Collection
efficiency
m
92.23
81.52
80.42
80.11
81.75
86.60
95.01
95.92
96.95
77.74
78.35
86.25
69.86
76.12
88.91
85.84
90.50
93.80
79.22
79.21
69.36
62.00
57.57
60.69
61.49
47.90
38.74
7.56
32.61
54.84
75.88
Transfer
units
(V
2.555
1.689
1.631
1.615
1.701
2.010
2.999
3. 198
3.490
1.503
1.530
1.984
1. 199
1.432
2. 199
1.955
2.354
2.780
1.571
1.571
1. 183
0.9675
0.8573
0.9336
0.9542
0.6520
0.4900
0.0786
0. 3947
0.7951
1.422
ON
-------
TABLE 17. PERFORMANCE OF CONFIGURATION SS5 SPRAY SCRUBBER ON AEROSOL G
Test
code
SS5-G
1
2
3
4
5
6
7
8
9
10
Contactor
1 ine
size
(cm)
5.842
rate
(std m3/min)
2.102
1.634
0.9816
1.634
0.9816
4.795
4.268
3.653
3.195
3.953
Gas
velocity
in
line
(m/sec)
13.07
10.16
6.103
10.16
6.103
29.81
26.54
22.71
19.87
24.58
Water rate
(liter/min)
21.80
15.86
9.652
21.80
21.76
21.80
21.80
21.88
21.95
21.88
pressure
(atm)
13.61
6.804
2.041
13.61
13.47
13.47
13.61
13.61
13.61
13.61
LiQuid- to-
gas ratio
(liter/std m3)
10.37
9.709
9.833
13.35
22.17
4.547
5.108
5.988
6.870
5.535
Effective
loss
(cm WC)
-7.672
-2.931
-0.8620
-10.86
-14.14
15.60
9.654
5.000
0.6034
-2.931
Contacting power
(kWh/1000 m3)
Gas
PG
-0.2090
-0.0798
-0.0235
-0.2959
-0.3851
0.4250
0.2630
0.1362
0.0164
0.07983
Liquid
p
PL
3.973
1.860
0.5651
5.113
8.409
1.724
1.957
2.294
2.632
2.120
Total
p
PT
3.764
1.780
0.5416
4.817
8.024
2.149
2.220
2.430
2.649
2.041
1.
p
PL
105.6
104.5
104.3
106. 1
104.8
80.23
88.15
94.40
99.38
103.9
Aerosol
rate
(mg/min)
0.1396
0.1156
0.1134
0.1017
0.09706
0.09731
0.1145
0.09531
0.09789
0.1306
Aerosol
(mfi/std m3)
In
0.06641
0.07077
0.1155
0.06229
0.09888
0.02029
0.02684
0.02609
0.03064
0.03303
Out
0.04511
0.06002
0.1145
0.03953
0.05482
0.01619
0.02221
0.02078
0.02425
0.02462
efficiency
(%)
32.08
15.20
0.904
36.53
44.56
20.23
17.23
20.34
20.85
25.46
Transfer
units
t
0.3868
0.1649
0.00908
0.4546
0.5899
0.2260
0.1891
0.2274
0.2339
0.2938
-------
TABLE is. rERFcr^A!:CE cr S"P.AV- cuincr scr.us"!:" ON .".LROGCL D
00
Test
code
SO-D
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Contactor
orifice
size
(cm!
3.81
2.54
Gas flov
(std nrVmin)
4.074
3.681
3. 214
2.778
2. 234
1.592
4.064
3.529
2.907
2. 326
0.6178
1. 397
4.024
3. 540
2.907
2.001
1.133
0.5273
3.953
3.423
2.893
1.883
0. 1800
0 . 7808
2.413
2.386
2.375
2.362
2. 180
2. 180
2. 132
2.122
1.894
2.032
1.792
1.745
Gas
velocity
in
(m/sec)
59.55
53.81
46.99
40.60
32.66
23.27
59.41
51.59
42.50
34.01
9.032
20.42
58.82
51.75
42.50
29.25
16.57
7.708
57.73
50.04
42.29
27.53
2.632
11.41
79.38
78.48
78.13
77.68
71.69
71.69
70.11
69.79
62.31
66.84
58.96
57.40
(liter/mini
20.63
20.63
20.63
20.63
20.63
20.63
18.32
18.24
18.24
18.19
18.24
18.24
15.22
15.22
15.22
15.22
15.14
15.14
11.32
11.32
11.32
11.32
11.32
11.32
20.70
18.13
15.06
11.32
20.70
18.13
15.37
11.32
20.70
18.13
15.22
11.32
(atro}
13.61
13.61
13.61
13.61
13.61
13.61
10.21
10.21
10.21
10.21
10.21
10.21
6.804
6.804
6.804
6.804
6.804
6.804
3.402
3.402
3.402
3.402
3.402
3.402
13.61
10.21
6.804
3.402
13.61
10.21
6.804
3.402
13.61
10.21
6.804
3.402
(liter/std m3)
5.064
5.604
6.417
7.427
9.234
12.96
4. 508
5. 170
6.275
7.818
29.53
13.06
3.781
4.298
5.234
7.604
13.36
28.71
2.863
3.306
3.912
6.009
62.87
14.49
8. 579
7.599
6.342
4.792
9.499
8.318
7.209
5.334
10.93
8.922
8.489
6.485
Effective
loss
fen WC1
46.63
34.39
21.81
13.06
3.879
-4.698
49.31
33.53
19.22
5.344
-9.913
-3.491
52.06
37.50
22.71
6.939
-2.370
-6.379
55.17
38.36
25.77
8.534
-3.362
-1.198
148.7
158.5
157.9
157.5
121.9
121.9
121.9
121.9
78.57
78.57
78.57
78.57
Contacting power
(kWh/1000 cm3)
Gas
PG
1.270
0.9369
0.5941
0.3557
0.1057
-0.1280
1.343
0.9134
0.5236
0.1456
-0.2700
-0.0951
1.418
1.021
0.6187
0.1890
-0.0646
-0.1738
1.503
1.045
0.7021
0.2325
-0.0916
-0.0326
4.052
4.317
4.302
4.291
3.321
3.321
3.321
3.321
2.140
2.140
2.140
2.140
Liquid
PL
1.940
2.147
2.459
2.845
3.538
4.965
1.295
1.486
1.803
2.246
8.484
3.753
0.7243
0.8233
1.003
1.457
2.559
5.500
0.2742
0.3166
0.3746
0.5756
6.021
1.388
3.287
2. 183
1.215
0.459
3.639
2.390
1.381
0.5108
4. 187
2.564
1.626
0.6211
Total
PT
3.210
3.084
3.053
3.201
3.643
4.837
2.638
2.399
2.327 '
2.392
8.214
3.658
2.143
1.845
1.621
1.646
2.494
5.327
1.777
1.362
1.077
0.8080
5.929
1.356
7.338
6.501
5.517
4.750
6.960
5.711
4.702
3.832
6.327
4.704
3.766
2.761
%
PL
60.43
69.62
80.54
88.89
97.10
102.6
49.09
61.92
77.49
93.91
103.3
102.6
33.81
44.63
61.84
88.51
102.6
103.3
15.43
23.26
34.79
71.23
101.5
102.4
44.79
33.59
22.02
9.66
52.29
41.85
29.37
13.33
66.18
54.50
43.17
22.49
Aerosol
rate
CmE/mlnl
4.498
4.496
3.890
3.103
3.894
3.836
4.037
3. 105
3.894
3.347
3.747
3.785
3.812
3.828
3.738
3.483
3.356
3.719
3.660
3.489
4.059
4.234
2.129
4.207
4.342
4. 346
4. 152
3.913
4.096
4.036
3.893
3.966
4.341
5.008
4.391
4.240
Aerosol
(mfi/st"^ m3^
In
1. 104
1.221
1.210
1.117
1. 743
2.410
0.9935
0.8801
1. 339
1.439
6.065
2.710
0.9475
1.081
1.286
1.741
2.961
7.054
0.9261
1.019
1.403
2.248
11.82
5.388
1.799
1.822
1.748
1.657
1.879
1.852
1.827
1.869
2. 292
2.465
2.450
2.430
Out
0.06054
0.09157
0. 1581
0.1831
0. 2443
0. 1693
0.06837
0. 1497
0. 2443
0.3238
0. 1804
0.3238
0.06235
0. 1250
0.2590
0.5642
0.6289
0.4994
0.06386
0. 1331
0.2961
0.9988
2.043
2.020
0.00994
0.01268
0.01117
0.01268
0.01542
0.01753
0.01934
0.02084
0.03012
0.04217
0.05241
0.06295
Collection
efficiency
(•/.)
94.52
92.50
86.93
83.61
85.98
92.98
93. 12
82.99
81.76
77.49
97.03
88.05
93.42
88.44
79.85
67.59
78.76
92.92
93.10
86.94
78.90
55.57
82.72
62.52
99.45
99.30
99.36
99.23
99. 18
99.05
98.94
98.88
98.69
98.29
97.86
97.41
Transfer
units
-------
TABLE 19. PERFORMANCE OF SPRAY-ORIFICE SCRUBBER ON AEROSOL G
Test
9?d£
SO-G
1
2
3
4
5*
6*
7
8
9
10
11*
12*
13
14
15
16
17*
18
19
20*
21
22
23
24*
25*
26*
27*
28*
29*
30*
31*
32
33
34*
35*
Contactor
orifice
size
(cm)
3.81
2.54
Gas flow
rate
(std m3/min)
3.963
3.540
2.914
2.254
0.7393
1.634
4.044
3.540
2.907
2.132
0.9816
1.403
3.993
3.540
2.907
2.082
1.328
1,145
3.932
3.552
2.900
1.872
0.809
2.405 '
2.379
2.356
2.370
2.122
2.157
2.141
2.122
1.850
1.857
1.799
1.781
Gas
velocity
in
orifice
(m/sec)
57.93
51.75
42.60
32.95
10.81
23.88
59.12
51.75
42.50
31.16
14.35
20.51
58.38
51.75
42.50
30.44
19.42
16.73
57.48
51.92
42.40
27.37
11.83
79.10
78.25
77.50
77.96
69.79
70.94
70.43
69.79
60.84
61.06
59.19
58.57
Water rate
(liter/mini
20.70
20.70
20.70
20.59
20.70
20.70
18.32
18.24
18.24
18.24
18.24
18.24
15.22
15.22
15.22
15.22
15.22
15.22
11.39
11.39
11.39
11.39
11.39
20.70
18.13
15.22
11.32
20.59
18.13
15.22
11.47
20.70
18.13
15.22
11.32
Water
pressure
(atml
13.61
13.61
13.61
13.61
13.61
13.61
10.21
10.21
10.21
10.21
10.21
10.21
6.804
6.804
6.804
6.804
6.804
6.804
3.402
3.402
3.402
3.402
3.402
13.61
10.21
6.804
3.402
13.61
10.21
6.804
3.402
13.61
10.21
6.804
3.402
Liquid- to-
gas ratio
(1'iter/std m3)
5.225
5.848
7. 104
9.134
28.00
12.67
4.530
5.153
6.275
8.559
18.59
13.00
3.810
4.298
5.234
7.307
11.46
13.29
2.898
3.208
3.928
6.085
14.08
8.609
7.621
6.457
4.775
9.704
8.406
7.106
5.405
11. 19
9.766
8.456
6.355
Effective
friction
loss
(cm WC1
46.55
29.91
15.69
3.362
-14.05
-4.741
49. 13
33.92
18.96
5.344
-7.413
-3.551
51.46
37.67
22.24
7.887
-0.8620
-2.414
55.17
41.72
25.75
8.448
-1.215
161.9
158.5
158.5
158. 1
121.9
121.9
121.9
121.9
81.28
81.28
81.28
81.28
Contacting power
(kWh/1000 m3)
Gas
PG
1.268
0.8148
0.4274
0.0916
-0.3827
-0.1291
1.338
0.9240
0.5166
0.1456
-0.2019
-0.0967
1.402
1.026
0.6058
0.2149
-0.0235
-0.0656
1.503
1.136
0.7014
0.2301
-0.0331
4.409
4.317
4.317
4.306
3.321
3.321
3.321
3.321
2.214
2.214
2.214
2.214
Liquid
?L
2.002
2.240
2.722
3.499
10.73
4.856
1.302
1.481
1.803
2.459
5.340
3.736
0.7299
0.8233
1.003
1.400
2.194
2.546
0.2775
0.3072
0.3762
0.5828
1.349
3.298
2.190
1.237
0.4573
3.718
2.415
1.361
0.5177
4.288
2.806
1.620
0.6087
Total
PT
3.270
3.055
3.149
3.591
10.35
4.726
2.640
2.405
2.320
2.605
5.138
3.640
2.132
1.849 .
1.608
1.614
2.171
2.480
1.780
1.444
1.078
0.8129
1.316
7.708
6.507
5.554
4.763
7.039
5.736
4.682
3.839
6.502
5.020
3.834
2.823
%
PL
61.22
73.33
86.43
97.45
103.7
102.7
49.30
61.58
77.73
94.41
103.9
102.7
34.24
44.52
62.33
86.69
101.1
102.7
15.59
21.28
34.91
71.69
102.5
42.79
33.65
22.27
9.60
52.82
42. 11
29.07
13.49
65.95
55.90
42.25
21.56
Aerosol
rate
(mR/min)
0.1303
0.1295
0.1230
0.1173
0.1070
0.1063
0.1077
0. 1078
0.09696
0. 1115
0.09507
0.08148
0.1153
0.09642
0.09722
0.08756
0.07431
0.08302
0.1014
0.0836
0.09314 '
0.09502
0.1053
0.1260
0.1237
0.1157
0. 1119
0.1376
0.1095
0.1170
0.1072
0.1245
0.1253
0.1067
0.1230
Aerc
(mR/s
In
0.03287
0.03658
0.04220
0.05204
0. 1447
0.06504
0.02664
0.03046
0.03335
0.05232
0.09686
0.05808
0.02888
0.02723
0.03344
0.04205
0.05595
0.07252
0.02579
0.02354
0.03211
0.05076
0. 1302
0.05239
0.05199
0.04910
0.04719
0.06484
0.05076
0.05463
0.05052
0.06729
0.06748
0.05931
0.06909
sol
ra tion
d m3)
Out
0.01755
0.02425
0.03261
0.03953
0.07861
0.04587
0.01709
0.02364
0.02703
0.04443
0.07628
0.05203
0.01732
0.02011
0.02703
0.03675
0.05407
0.05761
0.01596
0.01687
0.02771
0.04789
0.1087
0.00694
0.00881
0.00956
0.00956
0.01363
0.00836
0.01062
0.01273
0.02771
0.02914
0.01634
0.01837
Collection
efficiency
(%)
46.63
33.72
22.73
24.03
45.68
29.48
35.84
22.37
18.94
15.08
21.25
10.41
40.03
26.18
19.16
12.61
3.36
20.57
38.11
28.34
13.71
5.64
16.54
86.76
83.05
80.52
79.74
78.98
83.53
80.57
74.81
58.82
56.82
72.45
73.41
Transfer
units
(V
0.6279
0.4112
0.2579
0.2749
0.6103
0.3492
0.4438
0.2532
0.2100
0.1635
0.2388
0.1100
0.5113
0.3035
0.2127
0.1348
0.0342
0.2303
0.4798
0.3332
0.1475
0.0581
0.1808
2.022
1.775
1.636
1.596
1.560
1.804
1.638
1.379
0.8872
0.8397
1.289
1.325
VO
Aerosol unsatisfactory.
-------
V 0.02-0.30
0.33-0.56
0.57-0.76
A 0.77-0.89
O 0.90-0.99
CONTACTING POWER — kWh/1000 m3
FIGURE 28 CONFIGURATION SSI SPRAY SCRUBBER PERFORMANCE
ON AEROSOL D
70
-------
z
D
GC
111
<
cc
u.
O
IE
Ul
m
5
0.57-0.76
A 0.77-0.89
O 0.90-0.99
I I I I I I I
0.1
CONTACTING POWER — kWh/1000 m3
SA-4380-23
FIGURE 29 CONFIGURATION SS2 SPRAY SCRUBBER PERFORMANCE
ON AEROSOL D
71
-------
10
I I I I I I I
t-
z
c
u)
u.
g '
<
oc _
m
2
D
f
<^> 0.33-0.56
Q 0.57-0.76
A 0.77-0.89
O 0.90-0.99
CONTACTING POWER — kWh/1000 m3
10
SA-438O-25
FIGURE 30 CONFIGURATION SS3 SPRAY SCRUBBER PERFORMANCE
ON AEROSOL D
72
-------
in
C
UJ
(C
u.
o
CD
5
V 0.02-0.30
0.33-0.56
D 0.57-0.76
A 0.77-0.89
O 0.90-0.99
CONTACTING POWER — kWh/1000 m3
SA-438O-26
FIGURE 31 CONFIGURATION SS4 SPRAY SCRUBBER PERFORMANCE
ON AEROSOL D
73
-------
V 0-0.30
0.30-0.50
0.50-0.76
A 0.76-0.90
O 0.80-0.99
0.006
CONTACTING POWER — kWh/1000 m3
SA-4380-27
FIGURE 32 CONFIGURATION SS4 SPRAY SCRUBBER PERFORMANCE
ON AEROSOL G
74
-------
10
z
D
DC
IT
UI
CO
5
0.1
• 0 - 0.39
• 0.40 - 0.49
0 0.60 - 0.69
D 0.70 - 0.79
A 0.80 - 0.89
O 0.90 - 0.99
< 1.0
ORIFICE
SCRUBBER
0.1
1.0
CONTACTING POWER - kWh/1000 m3
10
SA-4380-51
FIGURE 33 CONFIGURATION SS5 SPRAY SCRUBBER PERFORMANCE
ON AEROSOL D
75
-------
3.0
1.0
cc
HI
li.
to
<
CC
LL
O
CC
HI
ffl
5
z
0.1
o.oi I
0.1
CONTACTING POWER — kWh/1000 m3
10
SA-4380-30
FIGURE 34 CONFIGURATION SS5 SPRAY SCRUBBER PERFORMANCE ON AEROSOL G
76
-------
10
I
t
z
D
cc
LU
<
CC
CC
LU
CD
1.0
0.1
1 I I I I I 11
ORIFICE
SCRUBBER
0 - 0.19
0.20 - 0.29
0.30 - 0.39
0.40 - 0.49
0.50 - 0.59
0.60 - 0.69
0.70 - 0.79
0.80 - 0.89
0.90 - 0.99
1.0
0.1
1.0
CONTACTING POWER — kWh/1000 m3
10
SA-43SO-31
FIGURE 35 SPRAY-ORIFICE SCRUBBER PERFORMANCE ON AEROSOL D
77
-------
10 >
CC
UJ
oc
li.
o
oc
UJ
CO
5
D
Z
1.0
0.1
0.05
• 0 - 0.39
• 0.40 - 0.49
V 0.50 - 0.59
0 0.60 - 0.69
Q 0.70 - 0.79
A 0.80 - 0.89
O 0.90 - O.S9
< 1.0
CONTACTING POWER — kWh/1000 m3
SA-4380-32
FIGURE 36 SPRAY-ORIFICE SCRUBBER PERFORMANCE ON AEROSOL G
78
-------
transfer units was plotted as a function of the total contacting power,
expressed in units of kWh/1000 m^. The baseline performance curves for
the orifice scrubber are shown in all the figures for reference. The
calculations of contacting power and conversion of units were made with
the following formulas:
Gas-Phase Contacting Power:
P. = 0.02724 F,, kWh/1000 m3 (1)
where ¥„ = effective friction loss, cm WC.
Liquid -Phase Contacting Power:
PT = 0.02815 p, (L/G) kWh/1000 m3 (2)
Lt L
where
p, = liquid feed pressure, atm gauge
r 3
G = gas flow rate, m /min
L = liquid flow rate, liters/min.
Total Contacting Power;
P = P + PT kWh/1000 m3 . (3)
1 (j L
Equation (1) is precise, because Fg is a measured value of the gas-
phase energy actually consumed in contacting. However, Equation (2) is
an approximation, since it gives the theoretical kinetic energy of the
water spray emerging from the nozzle. It does not take into account any
energy losses that may occur in the nozzle itself, nor any energy losses
that may be undergone by the emergent spray in ways that do not contribute
to gas/liquid contacting. Direct measurement of the loss in the nozzle
would be extremely difficult if practical at all; apparently few if any
such data have ever been obtained for spray nozzles. No direct measure-
ment of the energy losses of the spray that are not incidental to gas/
liquid contacting appears to be feasible at all. Consequently, the
influence of any such losses must be inferred from the performance of
the scrubber. Since some energy losses of these natures must take place,
it is reasonable to expect that liquid-phase contacting power will in
some degree appear inferior to gas-phase contacting power even if the
two forms are actually fully equivalent.
79
-------
Tes*:s of the Configuration SS4 and SS5 spray scrubbers and of the
spray-crifice scrubber were made with both Aerosols D and G; the remainder
of the spray scrubbers were tested only with Aerosol D.
There was always some gas friction loss attending the flow of the
air through the spray scrubbers. Hence, it was not possible to isolate
the effects on collection efficiency of liquid-phase contacting power
without -he interaction of gas-phase contacting power. Tests were made
over a wide range of conditions that gave widely varying ratios of liquid-
phase to gas-phase contacting power. It was immediately evident that the
liquid-phase contacting power and the gas-phase contacting power did not
have equal effects on collection efficiency. However, it was not im-
mediately evident how complex the factors in the performance deviation
might be.
In. .1 preliminary trial, the efficiency data were assembled in groups
according to values of a parameter f_, which is defined as the ratio of
the liquid-phase contacting power to the total contacting power. Since
there were few runs in which the values of f_ were precisely equal, it was
necessary to assemble groups of tests having reasonably narrow ranges of
f_ rather than discrete values of the parameter. Nevertheless, inspection
of the graphs (Figures 28-31) indicated that the deviations of the spray
scrubber data from the baseline orifice scrubber curve could probably be
correlatad as functions of f_ alone. In Figure 31, the type of relation-
ship is illustrated approximately by the three parametric curves corre-
sponding to groups of points having different average f_ values. Further-
more, scrubber Configurations SSI, SS2, and SS3 appeared to give
essentially the same performance, but Configuration SS4 gave much
poorer performance than the first three. The same type of behavior
appeared in tests of Configuration SS4 with Aerosol G (Figure 32).
Ir. the Configuration SS5 spray scrubber, the water jet produced
enough draft to maintain a flow of air through the scrubber without use
of the blower. Operated in this manner, the SS5 scrubber represented
the class of ejector-venturi scrubbers, which are popular for a variety
of ga.s-cleaning operations. Because the cocurrent, concentric nozzle
was located in a straight run of pipe, the water jet was less efficient
in producing draft than it would have been had it been located at the
inlet cf a venturi. However, in other respects the performance of SS5
should have been comparable to that of commercial ejector-venturi
scrubber,3. For a given water pressure and flow rate, the air flow through
the scrubber was determined by the resistance, and was controlled by vary-
ing the resistance to air discharge downstream of the scrubber. When the
scrubber was operated in this "ejector mode," there was always a net rise
in pressure across the scrubber.
The SS5 scrubber could also be operated with the blower on, increas-
ing the air flow to a level at which there was a net pressure drop across
the scrubber. In such a case, the draft was produced by both the blower
and the water jet. Where the scrubber was operated in the ejector mode,
the total contacting power was calculated by subtracting the power
80
-------
equivalent to the gas pressure rise from the total power supplied by
the water jet (see Tables 16 and 17). Typically, the total power supplied
in the jet was about 1.05 times the calculated total contacting power.
Nominally, the value of f_ was 1.0 in these cases, since all the contacting
power came from the water jet. However, some of this power dissipation
actually took place as gas-phase friction loss, which may have had a
significant effect on the performance of the scrubber. This point is
discussed in more detail below.
The performances of the Configuration SS5 scrubber on Aerosols D and
G are shown in Figures 33 and 34. The data points corresponding to
f_ = 1.0 were taken with the scrubber operating in the ejector mode. The
performance on Aerosol D was evidently very similar to those of Configura-
tions SSI, SS2, and SS3.
The tests with the spray-orifice contactor were designed to determine
whether the contribution of the liquid-stream energy could be improved by
close combination of the spray with the action of the gas-phase turbulence
promoter, the orifice. The 60° spray nozzle was positioned so that the
spray would cover the area of the orifice. The jet also produced enough
draft that the scrubber could be operated in the ejector mode, without
use of the blower. The test results on Aerosol D are presented in
Figure 35 and Table 18, and those on Aerosol G in Figure 36 and Table 19.
The value of the tests on Aerosol G was reduced because wide variations
in the aerosol (revealed by the signatures) required excluding many of
the data points (see Table 19).
With the spray-orifice scrubber, numerous runs were made at low
values of f_; the data points fell close to the baseline orifice scrubber
performance curve (Figure 35). However, for runs at higher values of f_,
the data generally agreed closely with those for the Configuration SS5.
Evidently, no advantage was gained by combining the action of the orifice
with that of the spray.
To permit a comparison of the different spray scrubber configurations
and the spray-orifice scrubber, the data were rationalized in a series of
graphs (Figures 37-45) in which the data for all the scrubbers correspond-
ing to given ranges of f_ were plotted together. For Aerosol D and
f_ = 0-0.39, the data points for all the configurations fell close to the
baseline curve (Figure 37). Some scatter could be expected because of
the relatively wide range of f_ values, although all the values were low.
It is notable that the data points fell along both branches of the
orifice scrubber performance curve included in the figure for reference.
This tends to support a view that the two-branched performance curve is
characteristic of the aerosol and not of the orifice scrubber itself.
At values of f_ in the range 0.40-0.59, the differences in scrubber
unit performance started to emerge (Figure 38). The data points for the
SS5 spray scrubber and the spray-orifice scrubber continued to lie close
to the orifice scrubber curve. Those for SSI and SS3 deviated substan-
tially, and those for SS4 deviated still more. The two additional per-
formance curves, one for SSI and SS3 and the other for SS4 (broken lines),
81
-------
10
z
I
cr
i-
(E
UJ
CD
ORIFICE
SCRUBBER
SCRUBBER
CONFIGURATION
O SS 1
0 SS 3
D SS 4
A SS 5
O SO
f = 0 - 0.39
1.0
CONTACTING POWER — kWh/1000 m3
10
SA-4380-33
FIGURE 37 PERFORMANCE OF SPRAY SCRUBBERS ON AEROSOL D AT
VALUES OF f = 0 - 0.39
82
-------
10
CC
HI
u_
U)
DC
o
X
111
CD
5
0.1
0.1
SCRUBBER
CONFIGURATION
< SS 1
0 SS 3
D SS 4
A SS 5
O SO
f = 0.40 - 0.59
ORIFICE
SCRUBBER
1.0
CONTACTING POWER - kWh/1000 m3
10
SA-4380-34
FIGURE 38 PERFORMANCE OF SPRAY SCRUBBERS ON AEROSOL D AT
VALUES OF f = 0.40 - 0.59
83
-------
10
z
I
10
i
cc
a:
01
CO
2
Z
0.1
0.1
SCRUBBER
CONFIGURATION
V SS 2
0 SS 3
D SS 4
A SS 5
O SO
f = 0.60 - 0.69
ORIFICE
SCRUBBER
1.0
CONTACTING POWER — kWh/1000 m3
10
SA-4380-35
FIGURE 39 PERFORMANCE OF SPRAY SCRUBBERS ON AEROSOL D AT
VALUES OF f = 0.60 - 0.69
84
-------
10
D
X
cc
I-
QC
LU
CO
5
D
Z
1.0
0.1
0.1
SCRUBBER
CONFIGURATION
< SS 1
V SS 2
0 SS 3
D SS 4
A SS 5
O SO
f = 0.70 - 0.79
ORIFICE
SCRUBBER
I I I I I I I
1.0
.CONTACTING POWER - kWh/1000 m3
10
SA-4380-36
FIGURE 40 PERFORMANCE OF SPRAY SCRUBBERS ON AEROSOL D
AT VALUES OF f = 0.70 - 0.79
85
-------
10
z
D
u.
C/l
11.0
CC
u.
O
CC
UJ
CO
D
Z
I I I I
0.1 '
0.1
SCRUBBER
CONFIGURATION
1.0
CONTACTING POWER — kWh/1000 m3
10
SA-4380-37
FIGURE 41 PERFORMANCE OF SPRAY SCRUBBERS ON AEROSOL D AT
VALUES OF f = 0.80 - 0.89
86
-------
10
I
(/}
i
D
DC
LU
U.
DC
a:
tu
CD
1.0
0.1
0.1
I I I I I I I I
SCRUBBER
CONFIGURATION
<
V
0
D
A
SS 1
SS 2
SS 3
SS 4
SS 5
O SO
f = 0.90 - 0.99
1.0
a
CONTACTING POWER - kWh/1000 m3
10
SA-4380-38
FIGURE 42 PERFORMANCE OF SPRAY SCRUBBERS ON AEROSOL D AT
VALUES OF f = 0.90 - 0.99
87
-------
10
I
t
z
IT
LU
U.
CO
<
oc
IT
UJ
CO
5
Z
1.0
ORIFICE
SCRUBBER
SCRUBBER
CONFIGURATION
A SS 5
O SO
f = 1.0
0.1
0.1
1.0
CONTACTING POWER — kWh/1000 m3
10
SA-4380-39
FIGURE 43 PERFORMANCE OF SPRAY SCRUBBERS IN EJECTOR
CONFIGURATION ON AEROSOL D
88
-------
1.0
I
(SI
U)
< 0.1
DC
o
o:
uj
CO
0.01
0.1
SCRUBBER
CONFIGURATION
D SS 4
A SS 5
O SO
f = 0.90 - 0.99
1.0
10
CONTACTING POWER - kWh/1000 m3
SA-4380-40
FIGURE 44 PERFORMANCE OF SPRAY SCRUBBERS ON AEROSOL G AT
VALUES OF f = 0.90 - 0.99
89
-------
10
2
I
2
tr
LLI
LL
OT
<
DC
h-
o:
LU
m
S
z
1.0
SCRUBBER
CONFIGURATION
A SS 5
O SO
f = 1.0
0.1 '—
0.1
CONTACTING POWER — kWh/1000 m3
SA-4380-41
FIGURE 45 PERFORMANCE OF SPRAY SCRUBBERS IN EJECTOR
CONFIGURATION ON AEROSOL G
90
-------
were arbitrarily drawn in parallel to the branches of the orifice scrubber
curve; the cumulative evidence of these data and those in the subsequent
Figures 39-45 indicates that the assumption of parallel curves is probably
reasonable.
At values of £ in the range 0.60-0.69, all the data points deviated
markedly from the orifice scrubber curve (Figure 39). No distinction
could be made among the data for SS2, SS3, SS5, and the spray-orifice
scrubber, but SS4 again gave distinctly poorer performance. The same
behavior appeared at f_ values of 0.70-0.79 (Figure 40). However, at f_
values of 0.80-0.89 a separation appeared between the group of data
points for SSI, SS2, and SS3 and the group for SS5 and the spray-orifice
scrubber (Figure 41). The same condition appeared, in accentuated
degree, at £_ values in the range 0.90-0.99 (Figure 42).
The performances of the SS5 and spray orifice scrubbers when operated
in the ejector mode (f_ = 1.0) are shown in Figure 43. Within the preci-
sion of the data there was no difference in the performances of the two
devices. Although the curves drawn in Figures 42 and 43 indicate that
the ejector configuration of the SS5 and spray orifice scrubbers gave
somewhat higher efficiency than the same devices at f_ = 0.90-0.99, it
is not clear whether the difference is actually significant.
The performances of the spray scrubbers on Aerosol G are similarly
presented for £ = 0.90-0.99 (Figure 44) and for the ejector configurations
(Figure 45). The results are generally qualitatively in accord with
those obtained with Aerosol D. However, the relative deviation of spray
scrubber data from the baseline orifice scrubber curves was markedly
greater for Aerosol G than for Aerosol D, which shows directly that the
relative inefficiency of liquid-phase contacting power, compared with gas-
phase contacting power, becomes greater as aerosol particle size decreases.
GENERAL DISCUSSION
Comparison with Previous Studies
In both the present investigation and the preceding one,6 the per-
formance of the orifice scrubber was demonstrated to be dependent (for
a given aerosol) only on the effective friction loss, and not to be
independently affected by gas velocity, liquid-to-gas ratio, or orifice
size. This finding supported earlier experience with other scrubbers and
other aerosols as well as reports by other investigators (see Appen-
dix A). However, still other investigators have continued to report
finding independent effects of gas velocity, liquid rate, method of water
injection, and other such variables. In some unpublished studies of
venturi scrubber performance, Semrau2 found instances of relatively
inferior performance at low liquid-to-gas ratios in the range of 0.1-
0.4 liters/m3 (1-3 gal/1000 ft^). The recent tests with the orifice
scrubber were always made with substantially higher liquid-to-gas ratios.
Other discrepancies in reported scrubber behavior are less readily
91
-------
accounted for, although the methods of testing may have been involved in
at least some instances.
In the unpublished studies,2 a bench-scale contactor that generally
simulated the Greenburg-Smith impinger gave the same performance as the
venturi and orifice scrubbers at low effective friction losses, but
showed increasingly inferior performance as the air jet velocity was
increased to increase the effective friction loss. Consideration of all
such experience led the present author to conclude that the performance
curve determined for a given aerosol with well-designed orifice and
venturi scrubbers (and presumably other designs as well) probably repre-
sents the maximum scrubber performance attainable, although certain
scrubber configurations and/or modes of operation may give poorer per-
formance.
Performance Curves
Tha present investigation, like the preceding one on "Wet Scrubber
Liquid Utilization,"6 has again affirmed the straight-line form of
scrubber performance curves, although revealing the added complexity of
two-branched curves. The basis of the straight-line form is empirical.
Although Calvert et al.8 have derived an equation for venturi scrubbers
that predicts a straight line having a slope of 1.0 for a monodisperse
aerosol, it is not obvious that a straight-line form should necessarily
hold for polydisperse aerosols in general.
The two-branched curves afford still less explanation, even if it
is assumed that the two branches are not actually composed of precisely
straight lines. Two-branched curves first appeared in the preceding
study with the same orifice scrubber. The ammonium fluorescein aerosol
used in that investigation was found to vary quite widely because of
the instability of the generator. In tests with ambient air as the gas,
two-branched curves appeared in varying degrees for some of the variant
forms of the aerosol, and not at all for one or two forms. No lower
branch appeared in tests at adiabatic saturation and condensing conditions,
but the appearance of a lower branch was very marked under vaporizing
conditions. In the present investigation, two-branched curves appeared
in studies of all the test aerosols except possibly Aerosol F. In addi-
tion, two-branched curves evidently appeared in tests of the spray
scrubbers., It is therefore evident that the two-branched curve is a
property of the aerosol, not the scrubber.
There is still no satisfactory general interpretation for the slopes
of the scrubber performance curves. Slopes less than 1.0, which have
most often been encountered in various investigations, indicate that
collection of an aerosol becomes progressively more difficult as higher
collection efficiencies are attained at higher levels of contacting
power. This is in accord with both theoretical and empirical evidence
that coarse particles are more readily collected than finer ones, at
least in the range in which inertial collection is dominant. It was
92
-------
hypothesized earlier that a precisely monodisperse aerosol should give
a performance curve with a slope of 1.0. There is little chance for an
experimental confirmation of this hypothesis, since production of an
adequately monodisperse aerosol on a scale adequate for scrubber testing
appears to be infeasible. Furthermore, slopes of approximately 1.0 have
appeared in curves for aerosols that were not even nominally monodisperse.
Finally, performance curves with slopes greater than 1.0 have appeared.
These appeared first in correlations of data from the literature,4'5 but
have now appeared in the present investigation and its predecessor.
In these last cases the lower branches have always had slopes greater
than 1.0. The single performance curve for Aerosol F had a slope
greater than 1.0, but greater data precision might have shown a two-
branched curve in this case, also.
A performance curve slope greater than 1.0 indicates that the aerosol
becomes more readily collectable as higher collection efficiencies are
attained with increasing contacting power. It is difficult to visualize
any single collection mechanism that would give such behavior. Inertia
gives preferential collection of larger particles and diffusion preferen-
tial collection of finer particles. Either mechanism operating alone
should presumably give a performance curve with a slope less than 1.0,
since operation at higher contacting powers and efficiencies would result
in preferential depletion of the more readily collectable fraction of
the aerosol. Possibly the steep lower branch of the performance curve
represents a region in which both diffusion and inertia were operative
but neither was strongly dominant. The lower branch did represent a
region of relatively low overall efficiency. On aerosols of the sizes
under study, the inertial collection efficiency was low in this contacting
power range, and presumably collection by diffusion was low in any case.
However, if both mechanisms were increasing in effectiveness in this
region of operation, it seems that some such behavior as was observed
might result. This would certainly depend on how the efficiency of each
mechanism was varying with contacting power and how these efficiencies
were varying with respect to each other.
The "knee" in the performance curve, or transition from the lower
branch to the upper branch, possibly represents the point (or narrow
region) at which inertial deposition reached such effectiveness that it
clearly dominated the overall collection efficiency.
The foregoing speculations are probably still too crude to be termed
a hypothesis. Nevertheless, projecting some such combination of mechanisms
seems necessary to permit visualizing a physical process that could yield
the observed phenomena.
Particle Size/Efficiency Relationships
The performance of the orifice scrubber is presented in Figure 46 as
a function of the aerosol mass-median particle diameter as determined
from the cascade impactor tests (Table 3). The number of transfer units
93
-------
10
(ft
I
CC
ill
li.
1.0
CC
UJ
m
0.1
0.1
I I
• FE - cm WC
I I
1.0
MASS - MEDIAN PARTICLE DIAMETER —
10
SA-4380-56
FIGURE 46 PERFORMANCE OF ORIFICE SCRUBBER AS A FUNCTION
OF AEROSOL PARTICLE SIZE
-------
is shown with values of the effective friction loss as parameters. The
performance data were cross-plotted from Figures 20, 21, and 23.
The curves in Figure 46 can be interpreted only in a relative way.
Because the aerosols are polydisperse and have fairly wide particle-
size distributions, their collection characteristics are not adequately
described in terms of the mean particle size. Particularly in the low-
efficiency range, the portion of an aerosol being collected will tend to
be comprised mostly of particles larger than the mean size (unless the
speculation advanced above is valid).
Series Scrubbing . . .
Although the early study by Lapple and Kamack showed some improve-
ment in performance when an orifice and a venturi contactor were used in
series, subsequent experience2' ' did not show any clear differences in
scrubber performance when combination contactors were used. Following
the study by Semrau and Witham,6 Cooper19'20 derived relationships pur-
porting to show that series scrubbing could give higher efficiencies on
a specific aerosol than single-stage scrubbing. However, Cooper's
derivations require unrealistic premises, and his conclusions would not
apply to any real situation (see Appendix A). Conceivably, a series con-
tactor might be found empirically to give performance that is better, the
same as, or (as was actually observed in the present study) poorer than
that of a single-stage contactor. However, there is nothing in the
previous experience1 7 to suggest .that series contacting should give
performance systematically different from that of single-stage contactors.
In the present study, the performance of the multiple-orifice series con-
tactor probably represents a case in which a specific contactor gives
performance inferior to the "maximum performance curve" established with
the orifice scrubber. It does not necessarily follow that all series
contactors or scrubbers in series should give inferior .performance.
Spray Scrubbing
The results of the present investigation have apparently disproved
the earlier hypothesis that liquid-phase contacting power is equivalent
to gas-phase contacting power, although earlier data appeared to support
it. However, it is still not clear whether liquid-phase contacting power
is actually inferior to gas-phase contacting power, or whether the apparent
difference is due only to the difficulty of evaluating the actual (rather
than the nominal) contacting power, as is discussed above. An additional
question is raised by the new data that showed that the deviation of spray
contactor performance from the performance curve established for gas-
phase contacting power decreased with an increase in aerosol particle
size. If.this trend holds with aerosols larger than those used in this
investigation, it may be that for coarser aerosols (perhaps 2-3p.m and
larger) there is relatively little difference in the effectiveness of gas-
phase and liquid-phase contacting power. If so, this would account for
95
-------
some of the earlier results that were obtained with relatively coarse
aerosols.
T'.iere. are, of course, some friction losses in the nozzles, although
the Ioss3£ appear likely to be fairly small except in Configuration SS4.
The construction of the SS4 nozzle indicates that the internal friction
may consun.e a substantial part of the input power, which may explain the
marked inferiority of the SS4 performance.
The rest of the spray scrubber configurations gave generally similar
performance, although Configuration SS5 and the spray-orifice contactor
SO defi.nitely gave somewhat better performance than SSI, SS2, and SS3.
The similarity of the performances appears quite remarkable considering
the extreme divergences in the natures of the sprays and in their
orientations with respect to the gas stream. The wide-angle spray nozzle
SS3, the 60° spray nozzle used in the spray-orifice scrubber, and the
30° injector nozzle used in SS5 all were equipped with internal spin
vanes, so that the stream of water emerging from a nozzle was rotating in
a plane normal to the axis. The streams of water thus possessed kinetic
energy due both to the velocity of translation and to the rotation of the
liquid. Tie 30° injector nozzle delivered a coarse spray, while the 60°
and wide-angle (about 100°) nozzles delivered successively finer sprays.
The actual angles of spray discharge were presumably modified by the air
flow.
The S!52 nozzle delivered a flat, fan spray normal to the direction
of gas flow. Since the spray entered the contacting line from one side,
the sheet of water was presumably largely broken up by the crossflow of
the air. On the other hand, the concentric perforated tube SSI delivered
high velocity solid jets of water that should have traversed the gas
stream and impacted almost normally against the contacting line wall.
Atomization probably took place mostly upon impact of the water jets
against the wall.
Presumably, any of the kinetic energy of a liquid spray that is not
converted ultimately into turbulence and heat in the gas stream will not
be effective as contacting power. If spray should collide inelastically
with a wall of the scrubber, and the kinetic energy should be converted to
heat in the. film on the wall, the equivalent energy would be lost for
scrubbing purposes. A coarse spray discharged through a gas stream should
logically te expected to have retained more of its initial velocity and
energy upon reaching a wall than would a fine spray, and thus tend to
lose mo:re of its energy at the wall. Unfortunately, there appears to be
no way to determine any such losses directly.
It has been suggested that kinetic energy losses on the contacting
line walls 3iay account for the apparent inferiority of the liquid-phase
contacting power. The overall inferiority of liquid-phase contacting
power would be consistent with that explanation (whether correct or not),
but the relative performances of the different nozzles are contradictory.
The SS5 spray (30°) should travel farther before reaching the wall than
96
-------
would the'Other solid-cone sprays, but its droplet size was large. The
nominal 60° spray angle of the SO contactor should be contracted by the
air flow through the orifice, and the spray should have been finer. If
anything, the SO spray might be expected to lose less of its energy to
.the walls than the SS5 .spray, but the two devices gave the same per-
formance. It would appear most logical for the SSI spray to give by far
the poorest performance of all the devices if wall losses are the dominant
influence, but it gave essentially the same results as did the SS2 and
SS3 sprays.
These contradictions do not exclude the possibility that wall losses
are a significant factor, but they certainly indicate that the phenomena
involved are not simply accounted for. The impact of sprays on the walls
is not necessarily inelastic—at least, not completely so. If the impact
of the SSI water jets on the contacting line wall was not elastic, it is
difficult to see how the contactor could perform as well as it did.
There is yet another factor involved in the performance of the
Configuration SS5 and the spray-orifice contactors. When the Configura-
tion SS5 and spray-orifice contactors were operated in the ejector mode,
there was a net rise in gas pressure across the scrubber, but there was
also a pressure drop (magnitude not determined) across the scrubber down-
stream of the section in which the kinetic energy of the water jet was
transformed into gas stream pressure. This gas pressure drop, or friction
loss, constituted gas-phase contacting power even though the original
source of the energy was the liquid stream and the nominal total contact-
ing power was computed as liquid-phase contacting power. Since this
gas-phase component of the contacting power should be more effective than
the apparent liquid-phase component, an ejector scrubber might be expected
to give higher collection efficiencies than a spray scrubber having
nominally the same liquid-phase contacting power but no ejector effect.
This phenomenon can account for part of the difference in performance
between the ejector scrubbers and the SSI, SS2, and SS3 scrubbers.
The same conclusion should hold true even in cases in which the
blower was operated in conjunction with the Configuration SS5 spray and
spray-orifice scrubbers. Although there was a net gas pressure drop
across the scrubber, part of the draft was supplied by the water jet, and
some of the nominally liquid-phase contacting power was effectively gas-
phase contacting power..
Presumably the droplet size of a spray should be a critical factor
in particle collection, yet there was no indication that it had a major
influence independent of the liquid-phase contacting power. It appears
that, parallel to the case of contactors that draw their contacting
power from the gas stream, the efficiency of spray contactors on a given
aerosol depends almost entirely on the liquid-phase contacting power,
with little influence of contactor design or geometry. Although the SS4
configuration appears to be a relatively distinct exception to this rule,
this may be explainable as a case in which the actual liquid-phase con-
tacting power is reduced by internal nozzle friction.
97
-------
If it i:5 generally true that the maximum attainable collection
efficiency depends only on liquid-phase contacting power, with spray
droplet size having no independent effect, then the same performance
should probably be attained with spray nozzles actuated by other means
than liquid -pressure—for example, by mechanical spray generators or
pneumatic or steam atomization. A pneumatic atomizer may add a substan-
tial volume of gas to that entering the scrubber. Whether the added
flow of gas Ln the gas-liquid contacting zone will have an appreciable
effect on, the aerosol collection is problematical.
98
-------
REFERENCES
1. Lapple, C. E., and H. J. Kamack. Performance of Wet Dust Scrubbers.
Chem. Eng. Prog. 51. (3), 110-121 (March 1955).
2. Semrau, K. T. Unpublished data. Stanford Research Institute,
Menlo Park, California, 1952-53.
3. Semrau, K. T., C. W. Marynowski, K. E. Lunde, and C. E. Lapple.
Influence of Power Input on Efficiency of Dust Scrubbers. Ind. Eng.
Chem. 50 (11), 1615-1620 (November 1958).
4. Semrau, K. T. Correlation of Dust Scrubber Efficiency. J. Air
Poll. Control Assoc. 10 (3), 200-207 (June 1960).
5. Semrau, K. T. Dust Scrubber Design--A Critique on the State of
the Art. J. Air Poll. Control Assoc. L3 (12), 587-594 (December
1963).
6. Semrau, K., and C. L. Witham. Wet Scrubber Liquid Utilization.
EPA-650/2-74-108, U.S. Environmental Protection Agency, Washington,
D.C., 1974. 126 pp.
7. Semrau, K. T., and C. L. Witham. Condensation and Evaporation
Effects in Particulate Scrubbing. Paper 75-30.1, Air Poll. Control
Assoc., 68th Annual Meeting, Boston, Massachusetts, June 15-20, 1975.
8. Calvert, S., J. Goldshmid, D. Leith, and D. Mehta. Scrubber Hand-
book, Vol. I. EPA-R2-72-1182, U.S. Environmental Protection
Agency, Washington, D.C., 1972.
9. Calvert, S., J. Goldshmid, and D. Leith. Scrubber Performance for
Particle Collection. AIChE Symposium Series, 70_ (137), 357-364
(1974).
10. Calvert, S. Engineering Design of Fine Particle Scrubbers. J. Air
Poll. Control Assoc., 24 (10), 929-934 (October 1974).
11. Boll, R. H. Particle Collection and Pressure Drop in Venturi
Scrubbers. Ind. Eng. Chem. Fundam. 12. (1), 40-50 (1973).
12. Behie, S. W., and J. M. Beeckmans. On the Efficiency of a Venturi
Scrubber. Can. J. Chem. Eng. 5_1 (4), 430-433 (August 1973)
13. Behie, S. W., and J. M. Beeckmans. Effects of Water Injection
Arrangement on the Performance of a Venturi Scrubber. J. Air Poll.
Control Assoc., 24 (10), 943-945 (October 1974).
99
-------
14. Stober, W., and H. Flachsbart. An Evaluation of Nebulized Ammonium
Fluorescein as a Laboratory Aerosol. Atmos. Environ. ]_ (7), 737-748
(July 1973).
15. Smith, W. B. Private communication. Southern Research Institute,
Birmingham, Alabama, 1976.
16. Rao, A. K., and K. T. Whitby. Nonideal Collection Characteristics
of Single Stage and Cascade Impactors. Presented to: 15th American
Industrial Hygiene Conference, Minneapolis, Minnesota, June 1-6, 1975.
17. Dzubay, T. G., L. E. Hines, and R. K. Stevens. Particle Bounce Errors
in Cascade Impactors. Atmos. Environ., 10 (3), 229-234 (1976).
18. Marple, V. A., and K. Willeke. Inertial Impactors: Theory, Design,
and Use. In: Fine Particles, B.Y.H. Liu, ed., Academic Press,
New York, 1976. pp. 411-446.
19. Cooper, D. W., L. W. Parker, and E. Mallove. Overview of EPA/IERL-
RTP Scrubber Programs. EPA-600/2-75-054, U.S. Environmental
Protection Agency, Washington, D.C., 1975. 112 pp.
20. Cooper, D. W. Theoretical Comparison of Efficiency and Power for
Single-Stage and Multiple-Stage Particulate Scrubbing. Atmos.
Environ., 10 (11), 1001-1004 (1976).
100
-------
APPENDIX A
LITERATURE REVIEW
The development of the "contacting power rule" is described in a
series of four papers;1"4 these papers also reviewed and correlated most
of the available data up to 1963 that were given in sufficient detail to
permit correlation. The general conclusion was that the performance of
a particulate scrubber could be correlated by an equation of the form
(1)
where P-p is the total contacting power as defined in the Glossary and
a and Y are empirical constants that are essentially characteristic of
the aerosol only. Data from a variety of sources2" confirmed the general
form of the relationship and lent much support to the equivalence of
different types of scrubbers in which the total contacting power was
derived from the gas stream in the form of gas pressure drop. Walker
and Walker and Hall made pilot plant and plant-scale tests of scrubbers
and reached the same general conclusions.
Since the late 1950s the contacting power rule has become widely
accepted among users and manufacturers of scrubbers. The physical
demonstration of this acceptance is the widespread production and use of
the variable-throat venturi or orifice scrubbers that are now produced,
in one form or another, by all the major scrubber manufacturers and by
some users as well. Much of the acceptance was based on general ex-
perience with pilot-plant or full-scale scrubbers operated on industrial
dusts, fumes, or mists. The results of such tests have commonly been
reported sketchily if at all.
There have been few attempts to carry out systematic comparative
studies of different devices. More commonly, investigations have been
concerned with a single device, sometimes with minor variations. The
results of a number of such investigations have been reviewed in a pre-
vious report.7 Discrepancies or outright contradictions among the
reported results of different investigations have been common. In some
of these instances, it has been obvious that the testing techniques have
been defective. At least some of the claims of high scrubber efficiency
on fine dusts with low power consumption have appeared because investi-
gators made tests with redispersed dusts that were still highly flocculated,
*
References for this appendix are collected at the end.
101
-------
but assumed that the effective particle size and size distribution were
the same at; given by some method for measurement of ultimate particle
size.
Becke:: et al.8 reported the results of a test-stand comparison of
about 14 commercial scrubbers of the low-, intermediate-, and high-
energy types whose power consumption could be measured in terms of the
gas pressure drop. Each was tested in its designed operating range.
All the te;3ts were made with a standardized slate dust, which was re-
dispersed with a compressed air ejector. The measured penetrations of
the different scrubbers fell in a reasonably narrow band when plotted
as functions of the gas pressure drop.
A similar testing program was carried on by Bayer AG and reported
at stages oy Wicke9 and Holter.10 Tests were made with dusts redispersed
with a compressed air ejector. However, most of the conclusions were
evidently drawn from fractional efficiencies based on particle-size
distributions determined by centrifugal sedimentation. Apparently,
samples of the dust fed and of that penetrating the scrubber under test
were analyzed by this technique. However, there is no reason to expect
that the dust fed was actually dispersed to the ultimate particle size
measured by the sedimentation method; hence, the absolute magnitudes of
the data reported—at least—are questionable. Some of the data make
it obvious that the dust fed was partly flocculated. It is not apparent
that even the effective particle size of the flocculated dust fed in the
tests of different scrubbers was always the same. Depending on the
scrubber design, a flocculated dust may be partly deflocculated in the
scrubber itself before it can be effectively contacted by the water, and
this may have significant effects on the performance.
The 28 different scrubbers tested by Bayer AG included some units
with mechanically driven gas/liquid contactors. The performances of the
different units were presented as functions of the power consumption
(presumably of all forms of power used in any given scrubber). Substan-
tial differences in performance were indicated for different devices,
but it appears uncertain (as noted above) just what degree of significance
these differences actually had.
Guntheroth11 made an experimental study of the Pease-Anthony venturi
scrubber and strongly asserted11'12 that the collection efficiency was
largely determined by gas velocity, liquid-to-gas ratio, and method of
water Injection, independent of gas pressure drop. Unfortunately, his
data were presented only in part, and not in forms to permit ready
extrac':ior for correlation on the basis of gas pressure drop. Hence,
the validity of Guntheroth's assertions is impossible to evaluate on the
basis of his own published data. It can only be noted that his conclu-
sions have, not been confirmed by the preponderance of other work, and
they have apparently not been repeated or applied since the original
investigation.
102
-------
Taheri and Raines13 investigated different methods of water injec-
tion into a pilot-plant venturi scrubber, using a synthetic methylene
blue test aerosol. They found a slight power saving (about 107o) with
water feed through a weir rather than through spray nozzles or a con-
tinuous slot. The scatter in the data appear to indicate that the aerosol
particle-size characteristics fluctuated considerably.
Taheri et al.14'15 studied the operation of a small scrubber with
orifice contactors and a butterfly-valve contactor. They used a synthetic
methylene blue aerosol, and measured the inlet and outlet aerosols with
four-stage cascade impactors. Collection efficiencies were measured for
each of the three size fractions (small, medium, and large particles) into
which the aerosol was classified by the cascade impactors. Unfortunately,
Taheri et al. chose methods of plotting the data that tended to obscure
the effects of the variables. In certain regions of operation, the
efficiency at a given pressure drop was found to vary with water rate,
orifice size, or the position of the butterfly valve. These independent
effects of the variables other than pressure drop tended to disappear at
higher water rates. It seems likely that the variations in behavior were
associated primarily with low liquid-to-gas ratios.
With both gas/liquid contactors, the scrubbers gave lower collection
efficiencies with the medium particles than with either the coarse or
fine particles. Taheri et al. concluded14 that the minimum efficiency
(at a particle size of about 0.8 yxm) was determined by the transition
from diffusion-controlled to inertia-controlled particle collection, but
the particle size seems unreasonably high for such an explanation.
Starting from Equation (1), Cooper16 derived relationships that
purportedly showed that if gas/liquid contacting is carried out in multiple
stages instead of a single stage, the collection efficiency obtained should
be greater or lower than that for the single stage, depending on the value
of Y- However, his derivations required assumptions that are not applicable
to any real system. Cooper illustrated his derivations with a convenient
case in which all the contacting power took the form of gas pressure drop.
He assumed two cases, one in which the values of a and Y are the same for
the single-stage scrubber and for each stage of the multiple-stage
scrubber, and another in which each stage of the multiple-stage scrubber
has the same values of a and Y but these values are different from the
ot and Y values for the single-stage scrubber. He then determined that
for the first case, the multiple-stage scrubber should give higher
efficiency than the single-stage scrubber for Y < 1 and lower efficiency
for Y > !• For the second case, the relationship was more complex.
Cooper also determined that the effect of the variation of Y should be
a maximum if each of the multiple contact stages of a scrubber had the
same pressure drop.
The essential deficiency in the derivations lies in the assumption
of constant values of ot and y in a series of scrubber stages. This
assumption is equivalent to a statement that each identical scrubber
stage has the same efficiency, or value of Nt. The values of a and Y
103
-------
are dependent on the character of the aerosol entering the scrubber whose
performance is defined by Equation (1). (This is the case regardless of
whether or not the values of Qt and Y are assumed also to be affected by
the scrubber design.) For ct and Y to remain constant from stage to stage,
the aeroisol must also remain constant, which can hold true only under one
of the following conditions:
(1) The aerosol is precisely monodisperse.
(2) The aerosol is polydisperse but the efficiencies on each
si.ze of particle are such that the particle-size distribu-
tion remains precisely constant from stage to stage.
The first condition is, for all practical purposes, an intellectual abstrac-
tion. The chance that the second condition will occur is infinitesimal,
and the situation would have no generality in any case. Thus, Cooper's
hypothetical cases do not represent any real conditions that could be
expected to exist.
In a r<;al series scrubber, the aerosol particle size and size
distribution will be altered during passage of the gas through each stage.
Thus, each, stage in succession will be presented with a "new" aerosol for
which the a and Y values will be different. Whether the overall efficiency
for any givan multiple-stage scrubber will be different from that of a
single-stage scrubber having the same pressure drop (or effective friction
loss) will aave to be determined experimentally. It cannot be assumed
that different series contactors in general will necessarily give effi-
ciencies consistently higher or lower than will be obtained with single-
stage contactors.
REFERENCES
FOR
APPENDIX A
1. Lapple, C. E., and H. J. Kamack. Performance of Wet Dust Scrubbers.
Chem. Eng. Prog. 51 (3), 110-121 (March 1955).
2. Semrau, K. T., C. W. Marynowski, K. E. Lunde, and C. E. Lapple.
Influence of Power Input on Efficiency of Dust Scrubbers. Ind. Eng.
Cham. 50. (11), 1615-1620 (November 1958).
3. Semrau, K. T. Correlation of Dust Scrubber Efficiency. J. Air
Poll. Control Assoc. 10. (3), 200-207 (June 1960).
4. Senran, K. T. Dust Scrubber Design—A Critique on the State of the
Art. J. Air Poll. Control Assoc. 13. (12), 587-594 (December 1963).
5. Walker, A. B. Enhanced Scrubbing of Black Liquor Boiler Fume by
Electrostatic Pre-Agglomeration: A Pilot Plant Study. J. Air Poll.
Control Assoc. 13 (12), 622-627 (December 1963).
104
-------
6. Walker, A. B., and R. M. Hall. Operating Experience with a Flooded-
Disc Scrubber—A New Variable Throat Orifice Contactor. J. Air Poll.
Control Assoc. 18_ (5), 319-323 (May 1968).
7. Semrau, K. T., and C. L. Witham. Wet Scrubber Liquid Utilization.
EPA-650/2-74-108, U.S. Environmental Protection Agency, Washington,
B.C., 1974. 126 pp.
8. Becker, H., H. Breuer, and L. H. Engels. Wet Scrubbers: Develop-
ment and Test Stand Results. Staub-Reinhalt. Luft (in English), 29_
(3), 8-13 (March 1969).
9. Wicke, M. Collection Efficiency and Operation Behavior of Wet
Scrubbers. In: Proceedings of the Second International Clean Air
Congress, Academic Press, New York, 1971. pp. 713-718.
10. Holzer, K. Erfahrungen mit nassarbeitenden Entstaubern in der
chemischen Industrie. Staub-Reinhalt. Luft, 34 (10), 360-365
(October 1974).
11. Guntheroth, H. Schwebstoff-Nassabscheidung aus Gasen mit dem
Venturi Scrubber. Fortschritt-Ber. VDI-Z Series 3, Issue 13 (1966).
12. Guntheroth, H. An Artificially Generated Aerosol for Wet Scrubbing
Tests on a Semi-Industrial Scale. Staub-Reinhalt. Luft (in
English) 28 (11), 82-87 (November 1968).
13. Taheri, M., and G. F. Raines. Optimization of Factors Affecting
Scrubber Performance. J. Air Poll. Control Assoc. 19_ (6), 427-431
(June 1969).
14. Taheri, M., S. A. Beg, and M. Beizaie. Gas Cleaning in a Wetted
Butterfly Valve. J. Air Poll. Control Assoc. 22. (10), 794-798
(October 1972).
15. Taheri, M., S. A. Beg, and M. Beizaie. The Effect of Scale-Up on
the Performance of High-Energy Scrubbers. J. Air Poll. Control
Assoc. 23_ (11), 963-966 (November 1973).
16. Cooper, D. W. Theoretical Comparison of Efficiency and Power for
Single-Stage and Multiple-Stage Particulate Scrubbing. Atmos.
Environ., 10 (11), 1001-1004 (1976).
105
-------
Appendix B
AEROSOL GENERATOR*
The main component of the aerosol generator was a modified Mist-02~
Gen Model EN145 Ultrasonic Nebulizer. The commercial device consists
of a ceramic piezoelectric crystal mounted in the bottom of a nebulizer
chamber. The liquid to be atomized is placed in the chamber and is in
direct contact with the crystal. A power source supplies radiofrequency
energy at 1.4 MHz to the crystal, whose vibrations produce cavitation in
the column of the liquid above the crystal. Fine droplets of the liquid
ejected frcm the liquid column are carried out of the chamber with an air
stream supplied for that purpose. The commercial use for the device is
the pre-oaretion of therapeutic aerosols in a droplet size range suitable
for inhalation therapy.1' 2
The b&sic aerosol generation system was modified in an SRI-supported
program in an attempt to stabilize the output and character of the
aerosol over long periods of operation. The original plastic nebulizer
chamber was replaced by a chamber consisting of a section of heavy-wall
glass tubing of 7.6 cm i.d. held between two stainless steel flanges.
The chamber was 26 cm high, and was designed so that the Mist-02~Gen
transducer unit that contains the piezoelectric crystal could be attached
directly to the bottom. The solution to be nebulized was supplied con-
tinuously to the chamber by gravity flow from an elevated reservoir. The
liquid level in the chamber was maintained at 7.5 cm by means of a vented
feed tube within the chamber.
Presaturated air entered the chamber tangentially, imposing a slight
centrifugal field on the aerosol stream being carried out at the top of
the chamber. The centrifugal field and a baffle at the exit of the
chamber served to remove coarse spray droplets from the aerosol stream.
The ultrasonic transducer delivers about 50 watts of power in the
atomizatio:.i process. Much of this energy is dissipated as heat and is
absorbed by the solution. Some of the heat is carried out of the
nebulizer by the air stream, but the temperature of the solution tends
to rise. Changes in room temperature can also affect the solution tem-
perature. Temperature changes may affect the operation of the generator
and the characteristics of the aerosol, directly or indirectly. A tem-
perature control system was therefore incorporated into the aerosol
generation system. The nebulizer chamber was insulated for protection
*
References for this appendix are collected at the end.
106
-------
against room temperature changes and air drafts. The solution in the
nebulizer changer was maintained at a fixed temperature by circulation
through a heat exchanger coil submerged in a constant-temperature bath.
A small diaphragm pump was used to pump the solution from the bottom of
the chamber, through the heat exchanger coil, and back into the nebulizer.
The bath temperature was maintained at 30°C throughout the study of
scrubber performance.
Since the size of a particle produced from a droplet after evapora-
tion, of the solvent depends on the concentration of the original solution,
concentration of the solution by evaporation will result in an increase
in particle size. If the air used to carry the aerosol out of the
nebulizer were dry initially, it would also carry out some of the sol-
vent as vapor, leaving the solution more concentrated. In the aerosol
generator system, this problem was avoided by preliminary saturation of
the air at the temperature of the solution. Compressed air was passed
through a small ejector where solvent was drawn into the air stream and
atomized. The mixture of air and solvent (air and water in the present
study) was circulated through a stainless steel coil immersed in the
same constant-temperature bath as was used to control the temperature
of the solution in the nebulizer. The liquid water was then separated
from the air in a small centrifugal separator and returned to the
humidifier solvent reservoir, and the saturated air entered the nebulizer
chamber. The ejector nozzle served as a critical-flow orifice for the
air, and the air rate was regulated by setting the upstream air pressure.
It was found that variations in the voltage and wave form of the
electric current supplied to the power source unit of the generator
affected the quality of the energy that was in turn supplied to the
piezoelectric crystal. Therefore, a harmonically compensated voltage
regulator was inserted in the power supply line to the generator power
source.
The nominal liquid nebulization rate was about 1 cm-Vmin, but the
actual rate varied from 0.5-1.5 cm^/min. The aerosol output rates
varied with the carrier air rate, but also varied for undetermined
reasons that apparently were associated with the operation of the trans-
ducer itself. Fortunately, the variations in aerosol generation rate
did not produce corresponding variations in the particle-size character-
istics of the aerosols generated.
The variations in the particle size and/or particle-size distribu-
tion of the aerosol, as shown by the aerosol signatures, were evidently
not correlated with generator temperature, generator power input, or
aerosol generation rate. The variations did not, in fact, appear to be
correlated with any observed influence. Neither did they appear to be
completely random in their magnitudes. Instead, certain relatively
discrete aerosol forms appeared with greater or lesser frequencies. In
a total of 49 determinations made on Aerosol D, nine reasonably identifi-
able replicated variations of the aerosol appeared, with the number of
replications of individual variant forms ranging from 2 to 17. Another
variant form appeared only once.
107
-------
The determination of replications was made by superimposing the
graphs of the aerosol signature curves. The judgment of replication
was based more on the coincidence of the complete curves than on the
agreement: of the total number of transfer units for the four-stage mini-
scrubber. Curves that appeared to differ by no more than probable ex-
perimental error were considered to be replicates. A considerable
number of axact replicates were encountered. The variant form of the
aerosol that appeared with the greatest frequency (357») is shown as the
middle curve in Figure 18 of the report defined by the solid circular
data symbols and broken line. This variant form and three other forms
differing only moderately from it appeared with a combined frequency of
63%.
REFERENCES
FOR
APPENDIX B
1. Christoforidis, A. J., J. F. Tomashefski, and R. I. Mitchell. Use
o:~ an Ultrasonic Nebulizer for the Application of Oropharyngeal,
Laryngeal, and Tracheobronchial Anesthesia. CHEST _5j> (6), 629-633
(June 1971) .
2. Stevens, H. R.s and H. B. Albregt. Assessment of Ultrasonic
Ne.bulization. Anesthesiology 2]_ (5), 648-653 (September-October
1963) .
108
-------
GLOSSARY
Aerosol signature curve: The scrubber performance curve produced by a
miniscrubber run on a given aerosol.
Actual conditions (gas): The actual temperature and pressure of a gas
at the point of reference.
Collection efficiency: The fraction (Ip or percentage (1001]) of an
aerosol entering a collection device that is removed from .the gas
stream by the device.
Contacting power: The power that is dissipated in fluid turbulence (and
ultimately as heat) in the gas and liquid phases during gas/liquid
contacting in a scrubber. It is commonly expressed in units of
power per unit of volumetric gas flow rate:
3 3
kW/(1000m /hr) or kWh/1000 m Metric units
3
hp/(1000 ft /rain) English units
Effective friction loss: The equivalent of contacting power, expressed
in terms of equivalent gas pressure drop (due to friction loss
only) across the gas/liquid contactor. In most cases, it is approxi-
mately equal to the actual gas pressure drop across a device that
takes all its contacting power from the gas stream. It is commonly
expressed in the same units as gas pressure drop:
cm WC (centimeters of water column) Metric units
Liquid-to-gas ratio: The ratio of the volumes of liquid and gas brought
into contact in a gas/liquid contactor. It may be expressed on the
basis of the gas volume either at actual or at standard conditions:
3
liters/m (liters of water per actual Metric units
cubic meter of gas)
3
liters/std m (liters of water per
standard cubic meter of gas)
Scrubber performance curve: A correlation curve expressing the func-
tional relationship between the number of transfer units and the
contacting power (or effective friction loss). This is most con-
veniently presented on a log-log plot.
109
-------
Scrubber pressure drop: The total drop in gas pressure across a gas
scrubber, whether due to friction loss or to kinetic energy changes.
It is commonly expressed as the equivalent fluid head loss:
cm WC (centimeters of water column) Metric units
Inches WC (inches of water column) English units
Standard conditions (gas): The selected reference temperature and
pressure for specification of gas volumes. In this report, the
standard conditions are 25°C (77°F) and 1 atm pressure.
Transfer unit: Defined by the relationship
Nt = In [1/(1 - 11)]
where Nt = number of transfer units and T] = fractional collection
efficiency.
110
-------
CONVERSION FACTORS
Gas Flow
1 m3/hr = 0.58857 ft3/min
1 m3/min = 35.316 ft3/min
3
1 liter/min = 0.035316 ft /min
Gas Velocity
1 m/sec = 3.2808 ft/sec
Liquid Flow
1 liter/min = 0.26418 gal/min
1 liter/hr = 0.0044030 gal/min
3
1 m /hr = 4.4029 gal/min
Liquid-to-Gas Ratio
1 liter/m3 = 7.4808 gal/1000 ft3
Pressure
1 atm = 1.0133 X 105 N/m2 (pascals)
1 cm WC = 0.39370 inch WC
1 cm WC = 97.968 N/m2 (60°F)
Contacting Power
1 kWh/1000 m3 = 2.2784 hp/(1000 ft3/min)
111
-------
Effective Friction Loss
1 cm WC = 0.39370 inch WC
1 cm WC = 0.02724 kWh/1000 m3
1 inch WC = 0.1575 hp/(1000 ft3/min)
Aerosol Concentration
1 mg/m3 = 4.3697 X 10"4 grain/ft3
112
-------
TECHNICAL REPORT DATA
(Please read fnuructions on the reverse before completing)
1. REPORT NO.
EPA-600/2-77-234
2.
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Energy Utilization by Wet Scrubbers
5. REPORT DATE
November 1977
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO,
Konrad T. Semrau, Clyde L. Witham, and
William W. Kerlin
9. PERFORMING ORGANIZATION NAME AND ADDRESS
SRI, International
333 Ravenswood Avenue
Menlo Park, California 94025
10. PROGRAM ELEMENT NO.
1AB012; ROAP 21ABJ-005
11. CONTRACT/GRANT NO.
68-02-2103
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final: 6/75-8/77
14. SPONSORING AGENCY CODE
EPA/600/13
is. SUPPLEMENTARY NOTES ffiRL-RTP project officer for this report is Dale L. Harmon, Mail
Drop 61, 919/541-2925.
16. ABSTRACT
repOrfc gjves results of &n experimental bench-scale investigation of
the comparative performance of particulate scrubbers that draw contacting power
for gas /liquid contacting from the gas stream and from the liquid stream. The
three synthetic polydisperse test aerosols used had similar particle-size distribu-
tions, but mass-median particle diameters of 1.05, 0. 68, and 0.42 micrometers.
The different contactors were compared on the basis of collection efficiencies at
given contacting power levels , using the performance of an orifice contactor as a
reference level. A series contactor, consisting of staggered multiple orifices in
series, gave essentially the same performance as the single -or if ice contactor at the
higher contacting power levels, but poorer performance at the lower power levels.
Six different pressure spray nozzles were tested, one in combination with a single
orifice. All configurations gave poorer performance than did the reference orifice
scrubber at the same total contacting power. All but one of the spray contactors
gave very similar performances at comparable conditions, despite radical differences
in spray nozzle designs and spray configurations.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
Air Pollution
Scrubbers
Contacting
Performance
Comparison
Dust
Aerosols
Air Pollution Control
Stationary Sources
Energy Utilization
Particulate
13B
07A
13H
14 B
11G
07D
18. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
125
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
113
------- |