EPA-60a/2-77-238
December 1977
Environmental Protection Technology Series
FINE PARTICLE COLLECTION
BY A FLUX-FORCE/CONDENSATION
SCRUBBER: PILOT DEMONSTRATION
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 Servicp
Springfield, Virginia 22161.
-------
EPA-600/2-77-238
December 1977
FINE PARTICLE COLLECTION
BY A FLUX-FORCE/CONDENSATION
SCRUBBER: PILOT DEMONSTRATION
by
Seymour Calvert and Shamim Gandhi
Air Pollution Technology, Inc.
4901 Morena Boulevard, Suite 402
San Diego, California 92117
Contract No. 68-02-1869
ROAP No. 21ADL-002
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
A pilot-scale demonstration of flux force/condensation
(F/C) scrubbing for fine particle control was carried out
on a secondary metal recovery furnace. Control of the entire
source effluent, a maximum flow rate of 200 Am3/min (7,000
ACFM) at temperatures up to 800°C, required both particle
collection and acid gas absorption. Demonstration plant per-
formance was consistent with the preceding laboratory bench-
scale and pilot-plant studies.
The nature of the source emissions entering the scrubber
depended greatly on the type of scrap wire being incinerated
to remove the insulation from the copper wire and other metal
scrap and on the operating conditions. A conventional high
energy scrubber would be incapable of controlling emissions at
a practically feasible pressure drop from anything but the
premium grade of scrap. An F/C scrubber system would be fea-
sible for the control of lower (more polluting) grade scrap,
the type of scrap which had more commercial importance.
The system was generally capable of about 90% to 95%
\
efficiency on particles with a mass median aerodynamic
diameter of 0.7 to 0.8 ymA (about 0.3 ym physical diameter
for particles with a density of 4.0 g/cm3). This efficiency
was achieved with a 68 cm (27 in.) W.C. gas phase pressure
drop. A conventional high energy scrubber without F/C effects
would require pressure drops of roughly 250 cm (98 in.) W.C.
for 90% and 535 cm (210 in.) W.C. for 95% particle collection
efficiency.
F/C effects are those which accompany the condensation
of water vapor from the gas and are generally caused by
111
-------
contacting hot, humid gas with colder liquid and/or by injec-
ting steam into saturated gas. Mathematical models have been
developed for predicting the F/C effects and for use in scrub-
ber system design. Agreement between the model predictions
and experimental results was good.
This report presents the F/C system design details,
experimental results, analysis of results, description of
mathematical models, design of an optimized system, cost
estimates, and recommendations for future research.
IV
-------
CONTENTS
Abstract j^
Figures vi
Tables xi
Abbreviations and Symbols x^v
Acknowledgement xvii
1. Introduction 1
2. Summary, Conclusions and Recommendations 4
3. Demonstration Plant Site Selection 10
4. Engineering Design 23
5. System Performance Testing 41
6. Experimental Results and Discussion 51
7. F/C Scrubber Performance Prediction Model 77
8. Engineering Analysis 95
9. Future Research Recommendations 123
References 133
Appendices 134
«
A. Process Data 134
B. Particle Size Data 137
C. Particle Size Distribution Plots 144
D. Impinger Data 156
E. Calculation of Particle Number Concentration From
Cascade Impactor Data 159
F. Example Calculation and Prediction of Fractional
and Overall Penetration 169
v
-------
FIGURES
Number
3-1 Flowsheet for the 1.4 m3/min pilot scale F/C
scrubber system 16
3-2 Schematic of 1.4 m3/min pilot scrubber 19
4-1 Flow diagram of F/C scrubbing system 28
4-2 Quencher unit of F/C demonstration system 29
4-3 Scrubber unit of F/C demonstration plant 30
4-4 Cooling tower of F/C demonstration scrubbing sysgem . . 31
4-5 General elevation, F/C scrubber demonstration system. . 32
4-6 Top view, F/C scrubber demonstration system 33
4-7 Quencher details 34
4-8 F/C scrubber details 35
4-9 Electrical schematic of F/C scrubber demonstration 36
system 36
4-10 Furnace crossover duct to F/C scrubber demonstration
system 37
4-11 Front view of F/C demonstration scrubber 38
4-12 View of F/C scrubber system and metals recovery furnace 38
4-13 Crossover duct from furnace stack to F/C demonstration
scrubber 39
4-14 View of secondary metals recovery furnace and charging
process 39
4-15 Sieve plate scrubber of F/C demonstration system ... 40
4-16 Quencher unit of F/C scrubber system 40
5-1 F/C scrubber system instrumentation sheet 42
VI
-------
FIGURES (continued)
Number Page
5-2 Modified EPA sampling train with ex-stack cascade
impactor. Scrubber outlet sampling system 45
5-3 Modified EPA sampling train with ex-stack cascade
impactor. Quencher inlet sampling system 46
6-1 Particle penetration versus aerodynamic diameter
for run 3 50
6-2 Particle penetration versus aerodynamic diameter
for run 4 60
6-3 Particle penetration versus aerodynamic diameter
for run 42 61
6-4 Particle penetration versus aerodynamic diameter
(scrubber only) for run 43 61
6-5 Particle penetration versus aerodynamic diameter
for run 56 62
6-6 Particle penetration versus aerodynamic diameter
for run 58 62
6-7 Particle penetration versus aerodynamic diameter
for run 59 63
6-8 Particle penetration versus aerodynamic diameter
for run 61 63
6-9 Particle penetration versus aerodynamic diameter
for run 62 64
6-10 Particle penetration versus aerodynamic diameter
for run 64 64
6-11 Particle penetration versus aerodynamic diameter
for run 66 65
6-12 Particle penetration versus aerodynamic diameter
for run 69 65
6-13 Particle penetration versus aerodynamic diameter
for run 72 66
6-14 Particle penetration versus aerodynamic diameter
for run 73 66
VII
-------
FIGURES (continued)
Number
6-15 Particle penetration versus aerodynamic diameter
for run 74 67
6-16 Particle penetration versus aerodynamic diameter
for run 75 67
6-17 Particle penetration versus aerodynamic diameter
for run 76 68
6-18 Particle penetration versus aerodynamic diameter
for run 77 68
6-19 Effect of calculated particle number concentration
on agreement between predicted and experimental for
0.5 ymA diameter particles 74
7-1 Generalized F/C scrubber system 79
j
7-2 Multiple plate F/C scrubbing system 81
7-3 Predicted particle condensation ratio (f ) as a
function of particle diameter P 87
7-4 Predicted particle condensation rates (f ) as a
function of liquid bulk temperature. . .^ 87
7-5 Predicted particle condensation ratio (f ) as a
function of gas phase (saturated) temperlture 88
7-6 Predicted particle condensation ratio (fp) as a *
function of particle number concentration 88
7-7 Predicted particle condensation ratio (fp) as a
function of liquid phase heat transfer coefficient . . 89
8-1 Flowsheet for F/C scrubber optimum design at metals
recovery furnace 100
8-2 Saturator for F/C optimum design 103
8-3 Predicted grown particle size distribution for F/C
scrubber optimum design 105
8-4 Condenser for F/C optimum design 106
8-5 Predicted penetration for venturi scrubber as designed
for optimum system
Vlll
-------
FIGURES (continued)
Number Page
8-6 Venturi scrubber for F/C optimum design ....... 109
8-7 Assumed cumulative scrubber entrainment versus drop
diameter for optimum system .............. Ill
8-8 Cooling tower of F/C optimum design at metals
recovery furnace ................... -
C-l Inlet and outlet size distributions for run 3 ..... 145
C-2 Inlet and outlet size distributions for run 4
(scrubber only) ................... 145
C-3 Inlet and outlet size distributions for run 7 ..... 146
C-4 Inlet and outlet size distributions for run 42
(scrubber only) ................... 146
C-5 Inlet and outlet size distributions for run 43
(scrubber only) ................... 147
C-6 Inlet and outlet size distributions for run 56 .... 147
C-7 Inlet and outlet size distributions for run 58 .... 148
C-8 Inlet and outlet size distributions for run 59 .... 148
C-9 Inlet and outlet size distributions for run 61 ... .149
C-10 Inlet and outlet size distributions for run 62 .... 149
C-ll Inlet and outlet size distributions for run 64 .... 150
C-12 Inlet and outlet size distributions for run 66 .... 150
C-13 Inlet and outlet size distributions for run 69 .... 151
C-14 Inlet size distribution for run 71 ., ........ 151
C-15 Inlet and outlet size distributions for run 72 .... 152
C-16 Inlet and outlet size distributions for run 73 .... 152
C-17 Inlet and outlet size distributions for run 74 .... 153
C-18 Inlet and outlet size distributions for run 75 ... .153
IX
-------
FIGURES (continued)
Number Page
____ ^^
C-19 Inlet and outlet size distributions for run 76 ... .154
C-20 Inlet and outlet size distributions for run 77 ... .154
C-21 Inlet size distribution for run 78 ........... 155
E-l Particle penetration versus aerodynamic diameter
for run 56 ...................... 163
E-2 Inlet and outlet size distribution for run 56 .... 168
F-l Scrubber penetrations for collection by inertial
impaction as computed from equations 7-18 and
7-19 for different operating configurations ...... 175
F-2. Predicted grown particle size distribution
for run 56 data .................... 175
F-3. Particle penetration versus aerodynamic diameter
for run 56 ......................
F-4. Prediction of overall penetration for run 56 using
graphical integration .............. ... 176
-------
TABLES
Number Page
1-1 F/C scrubbing studies by A.P.T. for EPA 2
3-1 Summary of test data from stack sampling metal recovery
furnace 15
3-2 Performance of 1.4 m3/roin pilot plant 20
3-3 Particulate characteristics at 1.4 m3/min pilot plant . 20
6-1 Total particulate loadings at furnace stack during
different stages of F/C scrubbing demonstration program 52
6-2 F/C scrubber demonstration plant operating configura-
tions 54
6-3 F/C scrubber demonstration plant system performance . . 55
6-4 Particulate and impinger data*, F/C scrubber demonstra-
tion plant 59
6-5 Comparison of experimental data versus predicted pene-
trations for F/C scrubber demonstration 72
6-6 Measured variables, definition, precision and test case
nominal values for total efficiency calculation .... 76
6-7 Measured variables, definition, precision and test case
nominal values for size distribution calculation ... 76
8-1 Design criteria summary 99
8-2 Process streams for F/C scrubber optimum design at
metals recovery furnace 101
8-3 Total equipment cost estimate for F/C optimum design. .115
8-4 Direct and indirect cost estimate for F/C optimum
design 116
8-5 Operating cost of optimum F/C design 117
8-6 High energy scrubber costs for regular no. 1 wire . . .120
XI
-------
TABLES (continued)
Number Page
-_„._.._„ „ ::- - - - i *^" —
8-7 Cost comparison for premium wire recovery 122
9-1 Major industrial particulate sources for which F/C
scrubbing is attractive 127
A-l Process data F/C scrubber demonstration 135
B-l Inlet and outlet sample particle data for run 3. . . .138
B-2 Inlet and outlet sample particle data for run 4
(scrubber only) 139
B-3 Inlet and outlet sample particle data for run 7
(scrubber only) 139
B-4 Inlet and outlet sample particle data for run 42
(scrubber only) 139
B-5 Inlet and outlet sample particle data for run 43
(scrubber only) 139
B-6 Inlet and outlet sample particle data for run 56 . . .140
B-7 Inlet and outlet sample particle data for run 58 . . .140
B-8 Inlet and outlet sample particle data for run 59 . . .140
B-9 Inlet and outlet sample particle data for run 61 . . .140
B-10 Inlet and outlet sample particle data for run 62 . . .141
B-ll Inlet and outlet sample particle data for run 64 . . .141
B-12 Inlet and outlet sample particle data for run 66 . . .141
B-13 Inlet and outlet sample particle data for run 69 . . .141
B-14 Inlet and outlet sample particle data for run 71 . . .142
B-15 Inlet and outlet sample particle data for run 72 . . .142
B-16 Inlet and outlet sample particle data for run 73 . . .142
B-17 Inlet and outlet sample particle data for run 74 . . .142
B-18 Inlet and outlet sample particle data for run 75 . . .143
XII
-------
TABLES (continued)
Page
B-19 Inlet and outlet sample particle data for run 76. . . .143
B-20 Inlet and outlet sample particle data for run 77. .. .143
B-21 Inlet and outlet sample particle data for run 78. . . .143
D-l Impinger data and results 157
E-l Particulate data F/C scrubber demonstration ..... .165
E-2 Inlet and outlet sample data for run 56 167
F-l Example calculations for prediction of overall penetra-
tion of F/C scrubber system (run 56) 174
Xlll
-------
LIST OF ABBREVIATIONS AND SYMBOLS
a - Interfacial area of bubble, cm
A - Cross sectional area of scrubber, cm2
p
c - Mass concentration kg/m3, g/cm3 or g/DNcm3
C - Dependent variable
C - Cumulative particle mass loading, g/DNm3
C - Total particle mass loading, g/DNm3
C' - Cunningham correction factor, dimensionless
C - Heat capacity of gas, cal/g-°K
d - Diameter, m or cm
d, - Diameter of perforation, cm
d - Mass mean diameter of particle, cm
d - Aerodynamic particle diameter, ymA^d (C! p )l'z
pa P P
d - Performance cut diameter (aerodynamic), ymA
d - Geometric mean particle diameter, pro or ymA
d - Sauter mean diameter, ym
D - Diffusivity, cm2/s
DNm3 - Dry normal cubic meter, at 0°C and 1 atm
E - Efficiency, fraction or %
f - Fraction of vapor condensing on particles, fraction
fy - Volume fraction of gas condensing, fraction
F - Foam density, ratio of clear liquid height to total
foam height
G - Gas flow rate, m3/s
h - Heat transfer coefficient, cal/s-cm - °C
H - Humidity, g/g
k' - Mass transfer coefficient, gmol/cm2-s-atm
L - Liquor flow rate, £/s
LM - Latent heat of vaporization cal/gmol
m - Mass, kg or g
mg - Sample weight, mg
xiv
-------
LIST OF ABBREVIATIONS AND SYMBOLS (continued)
M - Molecular weight, kg/kgmol or g/gmol
M - Cumulative mass loading collected on that stage and
those below, gm/DNm3
n - Particle number concentration,#/DNcm3
i
n - Particle number concentration #/cm3
N - Number of scrubber stages
p^ - Vapor partial pressure, cm H20
PBM " Mean Partial pressure of non-transferring gas, atm
P - Pressure, cm W.C.,dynes/cm2, kg/cm2, N/ra2 or atm
P, - Barometric pressure atm
P - Orifice pressure, cm HaO
Pt - Penetration, fraction or percent
Pt - Overall penetration, fraction or percent
Pt* - Constant penetration value for particles below about
0.5 pmA size, fraction
Pt - Penetration due to impaction in the saturator, fraction
3-
Pt, - Penetration due to impaction in the condenser, fraction
Pt - Penetration due to diffusiophoretic in the condenser,
fraction
Pt, - Penetration due to impaction in the stages after the
condenser, fraction
AP - Pressure drop, cm W.C. or atm
q1 - Mass H20 condensed/mass dry air, g/g
q - Mass H20 condensed on particles/mass dry air, g/g
Qp - Gas volumetric flow rate, m3/s
QT - Liquid volumetric flow rate, m3/s or &/s
Lt
r - Radius, cm or jam
r - Distance, cm
R - Ideal gas law constant (82 atm-cm3/gmol-°K)
S - Saturation ratio, atm/atm
S - Sensitivity of error dC/C to an error 3X
t - Time, s
T - Temperature, °K (or °C, where specified)
T. - Impactor temperature, °C
J\
xv
-------
LIST OF ABBREVIATIONS AND SYMBOLS (continued)
u - Velocity, m/s or cm/s
u, - Gas velocity in perforations, cm/s
u D - Diffusiophoretic deposition velocity, cm/s
V - Volume, m3 or cm3
VDGM "* Volume °f dry Sas read by gas meters, m3
W - Weight o£ dryer, g
o
X - Independent variable
y - Mole fraction, gmol/gmol
y - Mass HaO/mass dry air, g/g
Z - Distance in "Z" direction, cm or m
Greek
y - Viscosity, g/cm-s
ym - Micrometer (micron)
ymA - Aerodynamic diameter d (C1 p ) */*, ym (g/cm3)1/2
p - Density, kg/m3 or g/cm
Pm - Molal density, gmole/cm3
p - Particle density
a - Geometric standard deviation of-particle size distribu
g tion
E - Summation
Subscripts
G - Gas phase
i - Interface
i - In
i - Initial
i - Individual particle size
i - Inertial impaction
L - Liquid phase
o - Outlet
p - Particle
pa - Aerodynamic
t - Total
xvi
-------
ACKNOWLEDGMENTS
A.P.T. appreciated the perceptive and helpful critical
review of the demonstration program by EPA personnel con-
cerned with this project: Dr. Leslie E. Sparks and Mr.
Dale L. Harmon, under the direction of Mr. James H. Abbott.
Several members of the A.P.T. staff other than the
authors of this report have contributed significantly to
the demonstration plant program. They are:, Messrs. Nikhil
Jhaveri, Russell Lyon, Willard Roper, Chuck Nguyen, and
Ms. Verne McAdams,
xvn
-------
CHAPTER 1
INTRODUCTION
Flux force/condensation (F/C) scrubbing can, under the
proper circumstances, achieve high efficiency fine particle
collection at appreciably lower power input than conventional
high energy scrubbing. This advantage applies especially to
the particle diameter range between about 0.1 to 1.0 micron.
For smaller particles Brownian diffusion becomes an important
collection mechanism and for larger particles inertial
impaction is more effective.
F/C effects are those which accompany the condensation
of water vapor from the gas and are generally caused by con-
tacting hot, humid gas with colder liquid and/or by injecting
steam into saturated gas. The transfer of water vapor toward
the cold liquid surface sweeps particles with it and is referred
to as diffusiophoresis. Heat transfer from the gas to the
liquid also causes particle movement toward the cold liquid
and this is called thermophoresis. The condensation of
water on the suspended particles causes their mass (particle
plus condensate) to increase and this is referred to as particle
growth. The particles are easier to collect by inertial
impaction after they have grown by condensation.
The research and development series which preceded the
presently reported program was based on the prior scientific
and engineering literature, which was inadequate for engineering
design purposes. A summary of the EPA-sponsored contract pro-
grams carried out by A.P.T. is given in Table 1-1.
The demonstration plant study which is reported here
followed the laboratory pilot-plant program. The objective
of the demonstration was to test an F/C scrubbing system in
-------
Table 1-1
IX)
Title
F/C SCRUBBING STUDIES BY A.P.T. FOR EPA
Contract Report # § Date
Wet Scrubber System Study -
Scrubber Handbook (Vol. 1)
Final Report (Vol.2)
Feasibility of Flux Force/
Condensation Scrubbing
for Fine Particulate
Control
Study of Flux Force/
Condensation Scrubbing
of Fine Particles
Study of Horizontal Spray
Flux Force/Condensation
Scrubber
CPA-70-95
68-02-0256
68-02-1082
68-02-1328
Task # 10
NTIS #PB 213-016
(1972)
EPA 650/2-73-036
NTIS #PB 227-307
(1973)
EPA 600/2-75-018
(1975)
EPA 600/2-76-200
(1976)
Subject
F/C scrubbing principles
and potential discussed.
Theoretical analysis,
mathematical modeling,
computer predictions,
experimental data from
bench-scale F/C scrubbers
Laboratory pilot-plant
study of F/C scrubbing
in sieve plate and spray
apparatus.
Laboratory pilot-plant
study of F/C scrubbing
in horizontal spray
apparatus.
-------
an actual industrial setting and to obtain information on
performance and costs. A survey of prospective sources yielded
several candidates which were in the pilot plant (1975)
report. Detailed follow-up of these candidates led to the
selection of a secondary metals recovery plant, mainly on the
basis of its being the only available opportunity for a
definite commitment by the host company.
The following report presents the details of the demonstra-
tion plant design and the experimental program. A refined,
near-optimum design was made on the basis of the experimental
data and a revised design model. The process design and cost
estimates for F/C scrubbing are presented in comparison with
conventional scrubbing. As will be seen, this pollution source
cannot be satisfactorily controlled by conventional scrubbing
but can be by F/C scrubbing.
-------
CHAPTER 2
SUMMARY, CONCLUSIONS AND RECOMMENDATIONS
SUMMARY
Flux force and water vapor condensation effects enhance
fine particle scrubber efficiency and therby provide a means
for reducing air pollution control costs. Flux force/
condensation (F/C) scrubbing is applicable in situations
where the gas is hot or where low cost waste steam is avail-
able. A series of EPA contract research and development
programs proceeded from scientific principles to develop
the engineering basis for the study reported here: a pilot-
scale demonstration plant.
The objective of the demonstration was to test an F/C
scrubbing system in an industrial situation and obtain
information on performance and costs. Based on source test
data, the gas temperature, and the importance of this type
of source, a secondary metal recovery furnace was selected
for control in the demonstration. The furnace is used to
burn insulation material from various types of scrap wire
and the emissions are small particles of insulation compo-
nents, condensed metal chloride fume, hydrocarbons, and
halogen gases.
An F/C scrubbing system composed principally of a spray-
type quencher, a sieve plate column, a spray-type cooling
tower, and an induced draft fan was designed on the basis of
preceding phases of our research. A maximum flow of 200
AmVmin (7,000) ACFM) of flue gas was cooled and saturated by
a sodium carbonate solution spray in the quencher and then
further cooled to provide F/C effects in the 5-plate column.
During the program it became necessary to modify the system
in order to cope with corrosion of the hot gas duct and to
-------
increase particle collection efficiency to compensate for an
increase of source emissions.
The system was generally capable of about 90% to 951
efficiency on particles with a mass median aerodynamic
diameter of 0.7 to 0.8 ymA (about 0.3 ym physical diameter
for particles with a density of 4.0 g/cm3). This efficiency
was achieved with a 68 cm (27 in.) W.C. gas phase pressure
drop. A conventional high energy scrubber without F/C effects
would require pressure drops of roughly 250 cm (98 in.) W.C.
for 90% and 535 cm (210 in.) W.C. for 95% particle collection
efficiency.
Demonstration plant particle collection efficiency data
\vere in line with those obtained previously in our bench-scale
and laboratory pilot-plant experiments. A mathematical model
)
which is a simplified version of that previously developed
yielded predicted efficiencies which compared well with the
demonstration plant data. The revised model is much easier
to use than the old one and requires only a hand-held type
calculator for the prediction of F/C scrubber efficiency.
The uncontrolled source characteristics proved to be
exceedingly variable, depending on interactions of several
factors. The type of scrap being reclaimed changes the amount
and kind of insulation in an individual charge to the furnace.
Operator technique determines the size of a charge, furnace
temperature, air rate into the furnace and afterburners, and
the timing of operations. There is a large change in emission
rate as batch-wise burning cycle proceeds.
A refined F/C scrubbing system was designed on the
conservative basis of the worst anticipated source emissions
and utilizing the lessons learned in the demonstration program.
With the re-designed F/C system the control of particulate
matter from the incineration of premium no. 1 wire would
require a scrubber pressure drop of about 70 cm (28 in.) W.C.
-------
and a condensation ratio of about 0.3 g water vapor per g day
gas. The control of lower grade scrap emissions would require
about 115 cm (45 in.) W.C. pressure drop and a condensation
ratio of 0.3 g/g.
An optimized F/C system for the control of emissions
from the incineration of no. 1 regular grade wire would cost
$111,000 installed. The annual operating costs for one shift/
day based on 10-year straight line depreciation and electric
power at 4.5* kWh would be $24,100/yr. This optimized F/C sys-
tem would probably be capable of controlling emisssions from
no. 2 and 3 wire incineration, based on the limited amount of
test data available for operation with no. 2 and 3 wire.
A conventional high energy scrubber system to control
no. 1 regular wire incineration would be impractical because
of the extremely high pressure drop which would be needed. In
order to make a comparison between F/C and conventional scrub-
bing costs, two system designs were made on the basis of con-
trolling no. 1 premium wire incineration (although this would
be inadequate for commercial use). For this service a F/C
system would require a 70 cm W.C. overall pressure drop and
would cost $103,120 installed and $20,695/yr to operate. A
conventional system would require a 238 cm W.C. overall pres-
sure drop and would cost $111,240 installed and $31,160/yr to
operate.
-------
CONCLUSIONS
The following conclusions can be drawn from the results
of this study:
1. The performance of the demonstration plant was con-
sistent with previous results in bench-scale and pilot-plant
studies.
2. The plant as built was capable of controlling emissions
from no. 1 premium wire incineration within air pollution
emissions regulations. It was marginally adequate for no. 1
regular wire, depending on source operation conditions.
3. A conventional high energy scrubbing system would
require a pressure drop of at least 250 cm (98 in.) W.C. to
give the same efficiency as the F/C demonstration plant did
at 68 cm (27 in.) W.C.
4. Particle number concentration is very important in
F/C scrubbing because the more particles sharing a given
amount of condensing water, the smaller their final size. In
the metal recovery demonstration the particle number concen-
tration was extremely high, making it a severe test of F/C
scrubbing.
5. Source operation in terms of numerous parameters is
variable over a wide range and has a great influence on the
nature of the emissions to be controlled.
6. A design procedure which has been improved and
simplified from the previous version gives predictions which
are in good agreement with experimental results. The new
mathematical model is convenient to use and can be solved
with only a hand-held electronic computer.
7. The spray cooling tower performance was satis-
factory in the demonstration plant. Less power and a less
efficient entrainment (drift) separator would be required
if a filled cooling tower were used. Observation of
the solids behavior in the demonstration plant indicated
little or no problem with deposition on tank and tower
surfaces.
-------
8. Liquid waste disposal was no problem after the
scrubber liquid was neutralized with sodium carbonate.
9. The tube bank entrainment separator following the
scrubber performed satisfactorily.
10. Refinement of the F/C scrubbing system design after
the demonstration program yielded a more nearly optimum
design for the control of this source. An F/C scrubber capable
of controlling no. 1 regular wire incineration would cost
$111,000 to install and $24,100/yr to operate. The efficiency
requirement would be beyond the practical capability of a
conventional high energy scrubber.
RECOMMENDATIONS
Recommendations based on the experience gained in the
demonstration plant program are as follow:
A. Regarding design revisions for a secondary metals recovery
air pollution control system:
1. The source operation should be studied carefully and
modified so as to minimize the level of emissions and the
variation in emissions. Close control of charging and firing
practices is needed.
2. The saturator (quencher) should be located immediately
adjacent to the incinerator stack in order to reduce the
temperature and corrosion burden on the duct work.
3. High mass transfer efficiency is required to control
the acid gases so the total combination of saturator, condenser,
scrubber, and entrainment separator must be designed to provide
it.
4. A venturi scrubber appears to be the best choice for
particle collection after condensation and growth. Because
the mass transfer capacity of a venturi and its entrainment
separator is limited, the quencher and/or condenser must be
efficient for gas absorption.
B. Regarding future research and development on F/C scrubbing:
1. Laboratory research on particle growth in bench-scale
prototypes of condensers and scrubbers is necessary to provide
-------
data which can be used in validating and/or revising the
mathematical model. At present the model involves several
significant assumptions.
2. Field measurements of particle number concentration
and growth characteristics for a number of important air
pollution sources would provide the key information needed
to assess the suitability of F/C scrubbing for the specific
source. Apparatus of the same general size and nature as
that used for making performance tests could be taken
anywhere in the U.S. and operated by a 3-man crew.
3. Field studies of solids deposition on surfaces of
various materials which could be used for cooling tower
construction would give the most important piece of infor-
mation needed in selecting the type of cooler to use. These
tests would be done at the same sites as described in 2,
above, and could most efficiently be done concurrently with
the particle characterization tests.
4. The use of steam injection into saturated gas is an
attractive but insufficiently explored ramification of F/C
scrubbing. A bench-scale laboratory study could yield the
information needed to determine the optimum balance between
the quantity of steam to inject and the amount of condensation
by cooling. Engineering design studies followed by pilot
tests are needed to identify the best way of generating
steam inexpensively.
5. F/C scrubbing is highly suited to the control of fine
particles from basic oxygen furnace emissions and a demonstra-
tion plant would be valuable.
-------
CHAPTER 3
DEMONSTRATION PLANT SITE SELECTION
The main point of this study was to conduct a
pilot-scale demonstration of a flux force/condensation
scrubber system for collection of fine particulate emissions
from an industrial source. Thus, it was important to
find a company which operated a suitable plant and which
was willing to allow access to the plant for the pilot-
scale demonstration.
The following criteria were used for selecting the
industrial source:
1. The type of source must be classified nationally as
a major pollutant.
2. F/C scrubbing must be applicable to the source in
terms of the technical and economic feasibility. Ideally,
the pollutant gas for F/C scrubbing is characterized by hot or
humid conditions with a major portion of the particulates in
the submicron range.
*
3. The source should be either difficult or expensive
to control with presently available particulate control
devices.
A secondary non-ferrous metals recovery furnace was
selected based on the above criteria for selection of the
F/C demonstration scrubber. This chapter describes the
nature of the process and the characteristics of the
pollutants at the source. In addition, the results of a
1.4 m3/min pilot-plant setup at the site is discussed in
greater detail.
10
-------
Alternate Sites Studied
Several companies operating metal melting furnaces were
visited. During each visit, the demonstration program was ex-
plained and the specifics of a possible F/C scrubber system on
their emission sources were discussed. The pertinent informa-
tion on some of the alternate emission sources is described
below while the particulars on the actual site selected are
given in the next section.
1. Emission source: cupola furnace pair
"Existing control equipment: a spray chamber followed
by a baghouse
The exhaust gas was at about 958°C. From the cupola
the gas entered the spray chamber where 150 liters per minute
of water was sprayed in. The large particles were removed by
the water spray and the gas was cooled to 163°C. The gas
/
then entered the baghouse and was subsequently emitted to the
atmosphere. The total gas flow rate from a cupola was about
315 m:Vmin @ 163°C and the cupolas were used alternately.
The baghouse had frequent operating problems due to the
high gas temperature and the pressure drop through the bags.
The possibility of installing the F/C scrubber system down-
stream of the spray chamber was discussed with the company.
2. Emission source: electric arc furnace
Existing control equipment: baghouse
The arc furnace was charged every two hours with scrap
iron. The exhaust gas, about 200 m3/min at 140°C, entered
the baghouse through a 20-HP fan and was subsequently emitted
to the atmosphere. The particulates were mainly iron oxides
in the submicron range. The possibility of installing the
F/C scrubber system in place of the baghouse was discussed
with the company.
3. Emission source: metal sweating furnace (incinerator)
Existing control equipment: afterburner for unburned
gas.
The gas-fired incinerator was charged approximately every
20 minutes with utility electrical wires and some synthetic
11
-------
with utility electrical wires and some synthetic materials such
materials, such as PVC. There was no particulate control
equipment for the exhaust gas except for the afterburners.
4. Emission source: metal sweating furnace (incinerator)
Existing control equipment: afterburner for unburned
gas.
The incinerator burned propane and was continuously charged
by a conveyer arrangement with aluminum, zinc, lead, and
synthetic materials such as PVC. The exhaust gas flow rate
was about 142 m3/min @ 482°C. The particulates emitted are
fines, mainly metal oxides and carbon. The possibility of
installing the F/C scrubber system on these incinerators
was discussed with the owner.
DISCUSSION OF ACTUAL SITE SELECTED
The Metal Recovery Furnace
After consulting with the Project Officer, it was decided
to pursue the possibility of evaluating the demonstration
scrubber system on the secondary non-ferrous metal recovery
furnace (incinerator). An agreement was worked out between
A.P.T., Inc. and the company operating the furnace, detailing
the scope of participation and task responsibilities of the
parties. A description of the furnace operation and emission
was as follows:
Emission Source: Secondary non-ferrous metal recovery
furnace.
Existing Control Afterburners for un-burned gas and
Equipment: combustibles.
The gas-fired furnace was used for the recovery of copper
from utility electrical wires and the recovery of aluminum,
zinc, and lead by the metal sweating process from the corres-
ponding scrap material. Two gas-fired afterburners were
located immediately downstream of the furnace to control emis-
sions of unburned gas and combustibles. However, the after-
burners had not been adequate to control ^articulate emissions.
especially during the copper wire reclamation.
Previous emission data indicated that the particulate
loading was dependent on the charge being burned (maximum
12
-------
for copper wire reclamation, lower for metal sweating opera-
tions) and the^peration of the afterburners. The furnace
was charged approximately every 20 minutes. Particulate
emissions were also higher during the first few minutes of
a charging cycle. Thus, the stack conditions and emissions
varied significantly, as illustrated, from the following two
sets of emission data available at the time of site selection:
Type of operation: Copper wire reclamation
Run 1 Run 2
*—
1. Emission rate, g/Nnr @ 12% C02 0.17 0.5
kg/hr 0.15 0.38
2. Gas flow rate, actual m3/min 193 158
3, Stack Temperature, °C 859 782
4, Process load, kg/hr 1816 1816
Both the afterburners were operated during the runs.
The wire reclamation industry classifies the scrap wire
as follows:
No. 1 Premium - This grade contains mostly weather
stripping, such as power lines. 80% by weight of
charge is reclaimed as copper.
No. 1 Regular - This grade of #1 wire contains 60%
copper and 40% plastic.
No. 2 This grade contains mainly thin wires below 16
gauge with thick rubber, P.V.C., nylon, etc.
insulations.
No. 3 No. 3 wire contains mostly heavier wire than
16 gauge with much heavy insulation.
EMISSION DATA FROM STACK SAMPLING
Emissions tests were scheduled to determine particle
loading and size distributions at various furnace opera-
ting modes. The variables for the different operating modes
were reclamation of different grades of wire and furnace
operation with and without afterburners.
The method used for particle loadings and size distribu-
tions was a modified EPA Method No. 5 with an out-of-stack
13
-------
University of Washington Mark III cascade impactor in place
of the filter. The particulate loadings, along with the
total emission rates, are given in Table 3-1.
Preliminary testing of the non-ferrous metal recovery
furnace supplied information on the type of particulate, the
particulate size, and the nature of the exhaust products. The
sampling tests did show that the particles were very fine and
perhaps partly volatile in the 650°C gas stream in the furnace
exhaust stack, indicating that control of the source would be
*'
a demanding, if not extreme test, of the F/C scrubbing process.
Miniature Pilot-Scale Scrubber System
A multiple plate column with a capacity to scrub 1.4
mVmin was selected for a small scale study of the source in
order to evaluate the best F/C system for the specific charac-
teristics of the selected plant. A process flow sheet of this
pilot scrubber system is shown in Figure 3-1. The hot stack
gas was cooled and humidified in the quencher and then passed
through the F/C scrubber. The liquor system was designed such
that the temperature of scrubber inlet water can be controlled
in tank number 2 by selective mixing of stream 5 and 11. The
amount of vapor condensed in the scrubber was controlled by
the appropriate selection of the scrubber inlet water temperature.
A schematic drawing of the system is shown in Figure 3-2.
The pilot scale scrubber as assembled had the following dimen-
sions and specifications:
A. Probe and Duct to Quencher --
Function -- To draw 2.5 m3/min (90 ACFM) of stack gas at
650°C (1,200°F) isokinetically into the system from the furnace
stack and carry it without appreciable heat loss into the
quencher.
Construction -- Probe tip: 10 cm I.D., 60° sloped sharp
entrance. Gas duct: 6.4 cm I.D., insulated 10 m long steel
pipe.
B. Quencher --
Function --To cool 2.5 m3/min (90 ACFM) of stack gas at
650°C (1,200°F) down to 1.4 m3/min (50 SCFM) at 75°C (168°F)
14
-------
TABLE 3-1. SUMMARY OF TEST DATA FROM STACK
SAMPLING METAL RECOVERY FURNACE
Case
No.
1
2
3
4
5
Furnace Operation
and Load
Aluminum scrap
No afterburners
#1 wire
No afterburners
#1 wire
2 afterburners
#2 wire
2 afterburners
#3 wire
2 afterburners
Impactor Load
g/DNm3
0.0204
0.0243
0.492
0.172
0.306
0.686
0.988
1.61
Impinger Load
g/DNm3
—
—
1.31
1.40
3.16
3.50
dpg
ymA
0.93
0.90
0.48
0.47
—
0g
4.6
5.2
5.8
4.0
Emission Rate
Kg/hr
0.086
0.104
2.45
0.85
7.81
10.1
20.1
29.7
-------
FAN
STACK
GAS
TO 10
DRAIN •*—
QUENCHER
£APUMP 1
F/C
SCRUBBER
12
11
TANK 1
PUMP 2
Stream
No.
1
2
3
4
5
6
7
8
9
10
11
12
Stack gas
Saturated
gas
Outlet gas
Water
Water
Water
Water
Water
Water
Water
City water
City water
Temp .
(°C)
871
75
63
38
66
66
68
75
66
68
21
21
Volume Flow
m3/min
3.14
1.42
1.08
£/min
5.68
2.12
5.98
1.51
1.21
3.86
3.56
3.56
start-u
Flow
kg/hr
73.8
62. 7
341.0
127.0
359.0
90.6
72.6
232.0
214,0
214.0
p only
g vapor
g dry gas
0.40
0.18
—
—
—
—
—
Figure 3-1. Flowsheet for the 1.4 m3/min pilot scale F/C
scrubber system.
16
-------
and nearly saturated condition. This conditioning and cooling
was accomplished by direct evaporation of recirculated liquor.
_ <
Construction -- Overall size: 46 cm I.D., 69 cm high.
Flow: counter current. Spray requirements: 1.51 fc/min of
recirculated liquor sprayed evenly through 3 spray heads at
1,100 dynes/cm2 (160 psi) gauge. Holding tank: 56 cm diameter
by 51 cm deep, located 2 m below quencher with an open top.
The intake to the pump was 10 cm below the liquor level. Pump:
Piston pump with 370 watt motor, capacity of 8 5,/min at 14,000
dynes/cm2 (200 psi).
C. F/C Scrubber --
Function --To remove particulates from the gas by F/C
mechanisms.
Design -- A cylindrical four-sieve plate scrubber was chosen
for the 1.4 m3/min pilot test. Partially recirculated liquor
was introduced at the top plate and cascades down to the bottom
plate, maintaining cross flow on the sieve plates. Conditioned
gas from the quencher, introduced under the bottom plate, flowed
upward through the sieve plates.
Construction -- Overall size: 15.25 cm (6 in.) I.D., 183
cm (6 ft) high glass column. Liquid requirements: 5.7 £/min
of partially recirculated liquor introduced behind inlet weir
of top plate. Sieve plates: Four sieve plates with round perfo-
rations. The perforations were 0.318 cm diameter spaced tri-
angularly 0.95 cm apart. Inlet and outlet weirs on each plate
were 6.4 cm high by 12 cm wide. Entrainment separator: Fine
mesh screen placed across column. Pump: Centrifugal type with
250 watt mator, capacity of 22 5,/min. Fan: Direct drive,
centrifugal type with 1.5 kW motor, capacity of 1.2 cm3/min
at 64 cm of water. Holding tank: 56 cm diameter by 51 cm deep,
located under scrubber liquid level of 51 cm deep was maintained
by city water line with float control. The scrubber return line
was extended 46 cm below the liquid line.
D. Instrumentation --
Function --To provide data on the liquid and gas flow
rates, temperatures, pressures, arid composition. Measurement
17
-------
locations: The points in the system where the measurements were
made continuously during a run are shown in Figure 3-2.
Method -- Temperatures: Type T thermocouples for 0 to 100°C,
and type K for 100°C to 650°C, connected to a 24-point recorder.
Pressure: Bourdon tube gauges were used for the high pressure
measurements, Magnehelic gauges were used for low pressure
measurments. Volume flow: total volumes of liquid added or
removed from the holding tanks were measured by positive dis-
placement rotor-type meters. Liquid flow rate: Measured by
rotameters. Gas flow rate: Measured by orfice connected to
manometer. Particulate loading in gas streams: Loadings and
size distributions were measured by sampling the streams with
cascade impactors.
E. Liquor Conditioning --
Function --To condition the scrubbing and quenching liquor
such that it was suitable for dumping into the municipal sewer
line.
Method -- The liquor was treated in a batch process manner
at the end of a run.
DISCUSSION OF RESULTS FROM 1.4 m3/min PILOT TESTS
Results of the eight performance tests are presented in
Tables 3-2 and 3-3. Runs 1 to 3 were made when the furnace was
operated for aluminum sweating and the results from these runs
are not included. The remaining five runs were made during
copper wire reclamation (no. 1 regular grade).
Overall, when the impinger catch was included in deter-
mining penetrations, the quencher performance appeared to be
comparable or better than the scrubber performance. However,
when the results were compared for the collection of particu-
lates by excluding the impinger catch, the scrubber proved to
have a better performance as would be expected. Both devices
were efficient for the removal of the vapor phase of the stack
gas, but it should be noted that due to the small capacity of
the quencher requirement in the pilot tests, the quencher
behavior and design were not representative of a full-scale
system.
18
-------
T , P
g g
V V V \.
N N N N
T = gas temperature
T., = water temperature
P = water pressure
P = gas pressure
c>"
FW = water flow meter
V = water total
volume meter
T p p
w W w
o-
Pump
Scrubber
Quench Tank
N N N N
T
w
f
Tank
W
T p
g g
TW
TW
V V S. V X. V \. N.
£
i
T
T
I
V.
Water
Supply
Tank
Drain
V X V IX. X.
N N N
w
w
Pump
T , P
Figure 3-2. Schematic of 1.4 m3/min pilot scrubber.
-------
TABLE 3-2. PERFORMANCE OF 1.4 m3/min PILOT PLANT
K)
O
Run
No.
4
5
6
7
8
Loading mg/DNm3
Quencher
Inlet
1,566
1,150
774
1,383
Scrubber
Inlet
1,004
358
338
537
130
Scrubber
Outlet
437
201
179
303
46
Overall Pt
With
Impinger
Catch
27.9
17.4
23.1
21,9
Without
Impinger
Catch
31.9
47.2
45.0
43.3
Scrubber Pt
With
Impinger
Catch
43.5
56.1
53.0
56.4
35.4
Without
Impinger
Catch
44.3
56.7
77.9
67.3
74.9
TABLE 3-3. PARTICULATE CHARACTERISTICS AT 1.4 m3/min PILOT PLANT
Run
No.
4
5
6
7
8
qf
g H20/
g D.A.
0.06
0.067
0.073
0.088
Quencher Inlet
dpg
ymA
0.84
0.70
0.80
0.85
ag
1.1
1.7
1.2
1.1
Scrubber Inlet
dpg
ymA
0.70
0.60
0.55
0.53
0.44
°g
1.6
1.9
1.9
1.5
1.7
Scrubber Outlet
dpg
umA
0.60
0.50
0.45
0.53
0.40
°g
1.7
1.8
1.6
1.5
1.7
-------
Performance of the four-plate F/C scrubber was comparable
to the results obtained earlier for a 14.0 m3/min (500 CFM)
four-plate scrubber tested under a previous EPA contract
(Calvert et al., 1975). Due to substantial heat loss in
the small diameter gas line from the furnace stack to the
quencher in the pilot system, the condensation ratio, q',
(g H20 condensed/g dry air) in the F/C pilot scrubber was
low. For higher condensation ratio, the particle penetration
across the scrubber \vould be lower.
CONCLUSIONS FROM 1.4 m3/min PILOT TESTS
The pilot scale experimentation yielded information on
the source, the emissions, and the F/C scrubber design. These
may be summarized as follow:
1. Particulate and gaseous emissions from the furnace
change significantly with the nature of the charge and with
time within the charging cycle. However, the gas flow rate
and temperature remained constant during the charging cycle
and depend only on the nature of the charge.
2. Particulate loading and the concentration of the
corrosive condensable gas (consisting of metal chlorides,
HC1, fluorides etc., depending on the charge) were the
highest for #3 wire reclamation and decrease successively for
#2 and #1 wire.
3. Emissions from the recovery of aluminium, zinc, and
lead can be adequately controlled by afterburners. However,
the afterburners were inadequate to control emissions from
the copper wire reclamation, which presently exceed the control
standards. Due to the amount of copper reclaimable (~80%) from
the no. 1 wire and its availability, controlling emissions from
no. 1 wire reclamation with F/C scrubbing was assigned the highest
priority.
21
-------
4. During the wire reclamation operation, significant
amounts of corrosive acidic gases were emitted and appropriate
measures were used to cope with this condition. It was decided
to build the quencher with type 316 stainless steel and the gas
duct from the quencher to the scrubber of fiberglass-reinforced
plastic.
5. In order to absorb the acidic gases and to ease the
demands on materials of construction, sodium carbonate was added
to the quencher and the scrubber water to maintain the pH above
6. The quencher capacity was increased above the initial design
in order to assure adequate mass transfer capacity as well as
other benefits.
6. The mass loading as monitored during pilot plant test-
ing was somewhat higher than previously encountered. Therefore,
initial demonstration was made with five plates in the scrubber
rather than starting with four and then adding the fifth. Insu-
lation for the gas duct from the furnace to the quencher was used
to minimize heat loss so that a condensation ratio, q, of 0.2 to
0.3 can be maintained in the scrubber. These conditions were
expected to result in adequate particle collection in the scrubber,
based on the pilot system results and projections from the earlier
data.
22
-------
CHAPTER 4
ENGINEERING DESIGN
After conclusion of the pilot scale tests, the program
proceeded to the installation of the demonstration plant at the
metal recovery facility. The pilot scale experimentation had
yielded information on the source, the emissions, and the F/C
scrubbing system design. This section describes the overall
process design and the detailed mechanical features of the F/C
scrubbing demonstration plant.
PROCESS DESIGN
A flow diagram of the F/C scrubbing system along with the
major equipment items is given in Figure 4-1. The system was de-
Q
signed to scrub 198 Am /min (7,000 ACFM) of flue gas, this volume
being the maximum measured flow of gases from the furnace.
The gases were drawn from the furnace stack to the quencher
and through the scrubbing system by a fan located at the outlet
of the system. The gases were preconditioned in the quencher by
circulating water sprays to reduce the volume and to saturate the
gases with moisture. They -were then cleaned by a scrubber with
five perforated plates which was designed with provisions for
locating a sixth plate inside the shell. The cleaned gas was then
vented to the atmosphere through a stack.
Water from the quencher was drained to a holding tank and
recirculated to the quencher inlet. Liquor from the scrubber was
collected in another holding tank and sprayed in the cooling tower
which utilized ambient air for evaporative cooling. The outlet
liquid from the cooling towers was then recirculated to the scrub-
ber. City water was provided for the scrubber and cooling tower
holding tanks for make up and start up.
23
-------
Water used to quench and scrub the gas in the system was
treated periodically and chlorides in the gases were controlled
by the adjustment of pH of the scrubbing liquor with sodium car-
bonate as required. t
As indicated in Figure 4-1, the scrubbing system consisted of
three major vessels which included the quencher, scrubber, and
cooling tower. The primary function of each major vessel in the
F/C scrubbing system was as follows:
Quencher: The major objective was to cool 198 Am3/min (7,000
ACFM) of stack gas at 650°C (1,200 °F) down to 105 Am3/min (3,700
ACFM) at 71.0°C (160°F). This cooling was accomplished by evapor-
ating recirculated water. In addition, the quencher absorbed and
neutralized the acid gases which are emitted during the reclama-
\
tion of insulated wire. See Figure 4-2.
F/C Scrubber: The F/C scrubber served to remove particulates from
the stack gas and to absorb acid gas. The F/C scrubber had a
design capacity of 105 Am3/min (3,700 ACFM) and its major function
was to remove fine particulate emissions from the gas stream.
See Figure 4-3.
Cooling Tower: The purpose of the cooling tower was to reduce
temperature of the scrubber liquor with a flow rate of 570 £/min
(150 GPM) from about 59.4°C (139°F) to 40.6°C (106°F). This was
accomplished by means of evaporative cooling. See Figure 4-4.
MEACHANICAL DESIGN
The mechanical details of the F/C scrubbing system are
shown in the construction drawings which are presented in
Figures 4-5 through 4-10. Photographs of the physical system
are given in Figures 4-11 through 4-16. A brief description
on the mechanical design for each of the quencher major vessels
is given below.
Quencher
The quencher included a 1.52 m (60 in.) diameter and 3.7 m
(12') long vertical cylindrical spray chamber with co-current
24
-------
gas-liquid contact. The shell was assembled from rolled 316
stainless steel sheets, with stainless steel angle support.
The inlet and outlet ducting was 46 cm (18") diameter, and a
hollow cylindrical entrainment separator was built, packed with
1.6 cm pall rings, and placed at the bottom of the quencher
upstream of the outlet.
Initially, the quencher spray requirement was 57 £/min
(15.0 GPM) of recirculated liquor which was sprayed evenly
through spray manifolds at 2.8 kg/cm2 (40 psig). Later, the
spray flow rate was increased to 90 £/min (24 GPM) in order
to improve the collection efficiency of the system. A holding
tank measuring 1.2 m x 1.2 m x 60.0 cm located under the quencher
unit and a self-priming,centrifugal-type pump at 2.8 kg/cm2
(40 psig) was provided for this purpose. The inlet piping was
3.8 cm (1-1/2") I.D. while the internal arrangement included
a 3.2 cm square stainless steel tubing. The water inside the
quencher then drained through a 7.6 cm (3") I.D. pipe into the
cooling tower sump tank and eventually was recirculated.
F/C Scrubber
The F/C scrubber consisted of a cylindrical 2.3m (7'6")
diameter and 3.0 m (10') high, five-plate structure with
provisions for locating a sixth plate in a split-tower design.
The inside walls were coated with Epoxy paint for protection
against corrosion and leaks. The stainless steel sieve plates
were identical and had triangularly pitched 0.48 cm ('3/16")
perforations. (Note: The perforation size for plate? 4 and
5 were reduced to 0.32 cm (1/8") at a later stage of the project.)
The inlet and outlet weirs were 0.46 m (I16") long and 5 cm
(2") high.
The gas was transported from the quencher by a 46 cm (18")
diameter duct and introduced under the bottom of plate 1. The
gas flowed vertically upward through plates 1 to 3, down through
a rectangular compartment into bottom of plate 4, and then
vertically upward through plates 4 to 6. The gas then entered
the entrainment separator consisting of two tube banks, con-
structed from 1.9 cm diameter PVC pipe. It was eventually
25
-------
vented to the atmosphere by a 25-HP, belt-driven, centrifugal-type
fan which was capable of 142 Am3/min (5,000 SCFM) at 56.0 cm
(22") W.C.
The liquor requirements of the F/C scrubber included a
maximum of 56.0 £/min (150 GPM) water on plates 3 and 6. Two
self-priming, centrifugal-type pumps of 560 Jl/min (150 GPM)
capacity and 0.91 kg/cm2 (13 psig) head and a holding tank
measuring 1.8mxl.8mx0.61m(6' x6' x2f) was provided
for this purpose. Water was introduced at both plates 3 and
6 positions, and it cascaded down to the bottom plates 1 and 4,
respectively, maintaining cross flow on the sieve plates. The
inlet piping to the F/C scrubber was 7.62 cm (3")" I.D. while
the water was drained out from the unit into the holding tank
through a 15.3 cm (6") drain.
Cooling Tower
The cooling tower consisted of a 5.3mx 3.0 mx 2.7m
(17'6" x 10' x 9') counter-current spray chamber. Ambient air
was induced through the spray chamber by an air-foil, blade-type,
belt-driven, tube-axial fan with a 3.7 kw (5 HP) motor and
capacity of 650 Am3/min (23,000 SCFM). During the tests it was
found that a high efficiency entrainment separator was required
before the air could be vented to the atmosphere. The entrain-
ment separator consisted of 3 rows of wave-plate baffles with
spacing of 1.3 cm (1/2") and a wave amplitude of 2.2 cm (7/8").
Baffle rows were offset 0.65 cm (1/4") and baffle direction
in each row was also reversed.
The liquor to be cooled was sprayed from the top at a
rate of 560 £/min (150 GPM) at 2.8 kg/cm2 (40 psig) through a
spray manifold. This liquor was pumped from the scrubber
holding tank and utilized a self-priming, centrifugal-type
pump with a capacity of 662 £/min at 2.8 kg/cm2 (40 psig).
The cold water was then collected in a 3mx 2.7mx 0.46m
(10* x 9' x 1'6") holding tank under the tower and sprayed back
to the scrubber unit.
26
-------
Piping
Polyvinyl chloride (PVC) piping was selected for the F/C
scrubbing system, due to its superior resistance to acidic and
brine solutions. Chlorinated PVC piping was specified for the
quencher circuit due to the higher liquid temperatures. The
liquor piping system was based on the scrubber operating configu-
rations, while the layout was based on accessibility and eases
of measurement and control.
The flow control system with valves and bypass lines were
designed to provide for accuracy and protection from transients
and water hammer effects. Relay systems and level controls were
specified for the pump electrical wiring to protect against pump
cavitation, and to ensure that the hot furnace gas flow will be
cut off if the quencher pump failed.
27
-------
Cooling
Tower Tank
STREAM
NO.
1
2
3
1 4
; 5
! 6
; 7
8
9
i 10
11
12
13
14
COMPOSITION
Flue Gas
Gas Mixture
Gas Mixture
Ambient Air
Gas Mixture
Water
Water
Water
Water
Water
Water
Water
City Water
Make-up Water
TEMP .
°C
649
71.7
50.6
31.4
-
59.4
71.1
40.6
59.4
59.4
59.4
40.6
21.1
59.4
VOL. FLOW
Am3/min (G )
or fc/min (L)
198
100
71.8
850
-
59.0
43.3
568
568
1150
568
549
18.9
13.6
MASS
FLOW
kg/hr
4,400
5,384
4,480
58,700
60,000
3,540
2,600
34,100
34,100
69,000
34,100
32,900
1,140
818
ENTHALPY
(0°C BASE)
kcal/kg
226(d.g.)
2H(d.g.)
67.5(d.g.)
19
30.5
68.4
71.1
40.6
59.4
59.4
59.4
40.6
21.1
59.4
y
g water
g d.gas
0.07
0.31
0.09
0.011
0.033
-
-
_
-
_
-
_
_
-
Figure 4-1. Flow diagram of F/C scrubbing system.
28
-------
PLAN VIEW
Hot
Stack Gas
n
Water
Entrance
3.7m
A—•*-
Quencher
PLAN VIEW OF
SPRAY BANK
1.52 m
Diameter
Each bank
rotated 120°
20 cm
51 cm
28 cm
Level of Three
Spray Cross Members
ELEVATION
VIEW
J
Gas to Scrubber
Water Exit
Figure 4-2. Quencher unit of F/C
demonstration system.
29
-------
PLAN VIEW
Weir Section
Liquor Inlet
Liquor Inlet
Gas Outlet
Gas
Inlet
L
Sieve
PI atp #^
#2
#1
/
f
«
C
> 3
Sieve
Plate #6
#5
#4
Spray
-Manifold
n
I I
+
COperat
3.(
ing Drain
ELEVATION
VIEW
Figure 4-3. Scrubber unit of F/C
demonstration plant.
30
-------
r
2.7m
3.0m,
PLAN
Spray Nozzle
Bank Cross Members
. 3 m
•
|~« 1.8m ».
y\ Air
1 1 1 1 1 1 1 1 1 1 1 I 1
3 1 1 p
v Water +*
**^A Sprav /x/
s Ambient ^
j Air u
Holding Tank
Entrainment
Separator
±
i
50cm
T
.30cm
1.5m
ELEVATION
VIEW
i—f^ Water to Scrubber
Figure 4-4. Cooling tower of F/C
demonstration scrubbing system.
31
-------
[-0
•^"ff^f
AIR POLLUTION TECHNOLOGY,
,..„.,.. Inc. „„.,„,.
//•/
d II*I°H>KI at rrlcrinci'te"^!'r»fl>llhl4 *J ftil
Figure 4-5. General elevation, F/C scrubber demonstration system.
-------
Ol
Figure4-6. Top view, F/C scrubber demonstration system.
-------
i-& 3« ** X" 1
->,tvof>vr w&trfp op ' N
X Ih. Br{.[-r(T =( Air Pollution
T«k»lw. lot- ud li innS lutjict to tkc Condi mo
th*t It ihill
Figure 4-7.Quencher details.
-------
-.. 1.7- *°** \ ""'"'••
KBIICt: till I dr«"ini U tht properif ot »tr folluinc
iul dlipo»
-------
'LI
i y
-
a - .-a f ;
C ~£2OV!-
0 -(/„,..,. l»e- «.i.t.,.u
ELECTRICAL SCMMtTIC
AND cououn PLAN
KHLiNO ac4i.e
NMM W wn
na DHI
^ 27-4OI
WI1ICE: til. 4r«ii( l«
1 ikir PolUtiwi
, .
tin ii iktll «t to t.pn.4««J, »rlM.
vt*t dlnai*4 of, dlfvctt* ar iwUrrellT'
M"'** '""
Figure 4-9. Electrical schematic o£ F/C scrubber demonstration system.
-------
...
> •, •
WTTB0 jtfT£K 9& flff
HOIXUMO.
TO Cl
s/e'uitvx iS£u> —
>Z *EQU
l/-^
- S/& DlA « I't-i —
4 fFlt.
'
i
V " FICLO 0*
TBJJJJO. FL««
WSWU A4I4»JM 30-0*
r''J"t <•-.• . TW= "**-
^ ~*~T 1 l _,//(»«* two «s*J - ^
rmn I ~ S » ~" ' ' J r
-wnrcwsx ^ | „ L
f
•4 i r
?
i ty'mu. OOLWO BOLT
i : f£gift;*'m 11 *;«,«» C '
!„ ,i-^r~ (43. '*•«•«" X
$*«"»£• WM.W . . 4 , *»>«* y «WT6» L
] B 1. «rw**fTSf- 54*
~ — " -"-- '-^-j i tM-t- 4^1 '«fTf"TfFT
f - ^-ree, fvc^i^n^1 "1 ; » ^M %'^/ uutw '
&'s7L PIPE , i f&g MUTTS*
! .
«•>/<«»< • ' ' - «»:««. D£T^/
*££?"
1 ,'T* 1
ELEVATION • ,,. I_.T |
..*'"'"" * - • J^H,
- „ .-^ t — - ^" "T -^ d
Si ~ 1 \ "^1 ' _4._^^^"
\ V«OIA FiEl-D DRILL \,F/ \i 1
Y , 6-RCOD Jr.^/ ^4
« 1 * ' 1 1
|
«SK A v
' i-^> \-.J- VeTTMK_ElATE
R6F (%( * PL.C* " U SCAL.E: V4"=i'-O
gCAttEjl^"sl'-o"
~<-^ (~'^\3
\ \
fi" MkPPlE* -^ t#£eT J-,- \ \
COUPLING'S ^^. C*t*f- PI*"*1* \ ^
f— * ""• ' rn 4
sEssa T - ., «., i- ' - 1
1 ^ 0 \_
1 1 r\ "flf^ "' ' — "^ iiPf
1 . .. ^-^^--^==^^^'-^37p L
r-f- 1 T- 1 _^-^-^ p *wnr«*>w«
1 1 'i _rf^— -* REMOVE EXIfl AtJJ=N '
i__ , l^. — •** /&i09&'B0i.7-fienjK£ ' jgynr.
11 — wan rww fteuae, uw* wrsntttr \~'
SftNt
nrrov 4-^
«JT« fl«e tusuutTKw MOT
— i ' ^ ^^^^^£-'T*^tfSe GKIST. SUSSFf fLATSS
&»«« «W«r 1" "*' •!
*'"""" 5£CTVW x4-X
,. uV^ /T ^-^A^v^^f ^y£J SC4i-JE
cn^^T"^ ~^} /Sigfj«_'7jg_ft»*.
£-«£7o& . y...|-| -'J^
i /T\ ^n(i*-»«isus« ^___. , |*-» ^
V^y tnslnxuxf-, i |
«««-««n.«'^ 'j1 | VILM
- .x^Tt'L
,^•"'1^ >x^^i===i=^=
pnr^n --J
'^ £J%f- (?)
- i,: AIR POLLUTION TECHNOLOGY.
S» BLot» IBC. Cmtkf«nl>
I FURHACE CROSSOVER DUCT
tf/. REPUCEM&JT
PKUKt KO. ^7 APN0WU
I "wa s
l Srsy^Esss =r
. 1 °«n» •"' raa. na. im
— . 27-315 ™""~
lir»i:0''!?!.'SS'i!:5§i;[!S^!ii"'
w |T mjj NVIIMW wr'B '"^"r «>j «•* u •« w k* wtaitiwi « «n*U> M'"**
Figure 4-10. Furnace crossover duct to F/C scrubber demonstration system.
-------
•
Figure 4-11. Front view of F/C demonstration scrubber.
Figure 4-12.
View of F/C scrubber system and metals
recovery furnace.
38
-------
Figure 4-13. Crossover duct from furnace stack to F/C
demonstration scrubber.
Figure 4-14. View of secondary metals recovery furnace
and charging process.
39
-------
*--
Figure 4-15.
Sieve plate scrubber of
F/C demonstration system,
Figure 4-16. Quencher unit of F/C
scrubber system.
-------
CHAPTER 5
SYSTEM PERFORMANCE TESTING
After the installation of the scrubbing system, peripheral
equipment was installed for monitoring the different process
parameters and testing the performance of the system. On comple-
tion of this task, startup of the demonstration plant was ini-
tiated, and thereafter the system was evaluated by compiling pro-
cess and performance data under different operating modes. The
following information regarding the F/C demonstration plant is
described in this section of the report:
1. Instrumentation system of the demonstration plant.
2. Performance testing equipment.
3. Demonstration unit operating procedures.
4. Testing procedures.
5. Data analysis and calculations.
INSTRUMENTATION SYSTEM
Instrumentation for the scrubber system provided for measure-
ments of flow rates, temperatures, pressures and humidity in the
gas and vliquid streams as indicated in Figure 5-1.
Three venturi meters were fabricated out of polyvinyl chlor-
ide material and were used to measure flow rates in the scrubber
low pressure, scrubber high pressure, and cooling tower liquor
piping. Two fluid manometers with mercury as the indicating
fluid were used to measure the pressure drop across these ven-
turi meters.
Two rotameters with flow ranges of 0-150 Jl/min (0-40 GPM)
were installed and used for measuring flow rates in the liquid
lines going to the scrubber center duct and quencher spray mani-
fold. Rotameters were also used to monitor the liquor flow rate
of the sodium carbonate spray system and water spray
41
-------
Gas Flow L: Liquor Flowrate H: Humidity T: Temperature PL: Particle Loading
Liquor Piping G : Gas Flowrate PH.- PH P: Pressure SD: Size Distribution
f-O
I I I I I I I MM!
SCHP
PUMP
Figure 5-1. F/C scrubber system instrumentation sheet.
PH,
-------
inside the crossover duct from the furnace stack to the
quencher.
The gas flow rates were measured with standard pitot tubes.
These measurements were made at the inlet to the quencher and
at the inlet to the blower.
The gas temperature was measured with thermocouples,
utilizing type T (copper/constantan) at temperatures below
2QQ°C and type K (chromel/alumel) at temperatures above 200°C.
The thermocouples were wired to a multipoint, millivolt recorder
which was located in the scrubber system control room.
Water temperatures in the scrubber unit were measured with
copper constantan Ctype T) thermocouples. In the high pressure
liquid lines, dial thermometers were used for measurement of
the liquor temperature.
Pressures in the gas lines were measured with either
inclined, straight-well-type manometers or Magnehelic gauges.
This selection was based on the accuracy and sensitivity of
the measurement required. Pressure in the liquor lines was
measured by standard type pressure gauges.
The moisture content in the quencher inlet gas was deter-
mined from a wet and dry bulb thermometer measurements. The
scrubber inlet and outlet gas streams we're considered
saturated and the moisture content was determined from the
temperature of the gas.
PERFORMANCE TESTING EQUIPMENT
The performance of the demonstration plant was evaluated
by measuring the total particle collection efficiency of the
system by using a modified EPA Method 5 technique. In addition,
A.P.T. cascade impactors were used to measure the particle
size distribution in the gas stream and obtain the demonstra-
tion plant efficiency as a function of particle size. The
apparatus used is described as follows:
43
-------
1. Four identical particle sampling trains (.see Figures
5-2 and 5-3) were fabricated and utilized at the furnace stack,
quencher inlet, scrubber inlet, and scrubber outlet locations.
2. All the sampling equipment was operated with inlet
nozzles appropriately sized to give isokinetic sampling. In
the majority of the runs, a pre-cutter with an approximate
cut diameter of 6 ymA was used to remove both the larger
particle loading and the entrained liquid.
3. The sampling probes were constructed of 1.27 cm
(0.5") stainless steel tubing and installed in the sampling
ports of the scrubbing systems.
4. The total particulate loadings were measured by using
10.2 cm (4") ex-stack filter holders, while the particle size
distributions were determined by ex-stack A.P.T. cascade
impactors. The collection device in each case was followed
by 4 impingers for gas absorption and 1 silica gel dryer to
remove moisture.
5. The sampling probes and collection device were heated
with insulated heating tape controlled with variable trans-
formers to prevent water condensation in the system. The dry
gas sample flow rate was measured with a dry gas meter, a
rotameter, and an orifice meter in each train.
6. Impactor substrates and filters were weighed with
an analytical balance to the nearest tenth milligram. Tare
weights were taken after drying in an electrical oven and
desiccator. Glass fiber filter paper (Gelman type "A") was
used for impactor substrates and filters.
SCRUBBER SYSTEM OPERATING PROCEDURE
This section summarizes the procedure for the operation
of the F/C scrubber demonstration unit. For brevity, the peri-
odic routine maintenance functions are not included here, and
only the normal startup, operation,and shutdown procedures are
given.
44
-------
PRECUTTER
STAINLESS
STEEL
PROBE
STACK
WALL
A. P . T .
CASCADE
IMPACTOR
en
ORIFICE METER
DRY GAS
METER
SILICA
GEL
DRYER
IMPINGER TRAIN
- ROTAMETEF
VACUUM
GAUGE
VACUUM
PUMP
Figure 5-2. Modified E.P.A. sampling train with ex-stack cascade impactor,
Scrubber outlet sampling system.
-------
ft m A T •» r T T~I *~i r*« *~i rrtt-t -ri T
PROBE
STACK
WALL
PRECUTTER
A.P.T. CASCADE
IMPACTOR
SILICA
GEL
DRYER
IMPINGER TRAIN
ORIFICE METER
DRY GAS METER
VACUUM
PUMP
ROTAMETER
VACUUM
GAUGE
Figure 5-3.Modified E.P.A. sampling train with ex-stack cascade impactor,
Quencher inlet sampling system.
-------
During startup, initially all the liquor tanks were filled
with city water and then the water was circulated from the
scrubber to the cooling tower. Cold water from the cooling
tower was pumped back into the scrubber. Simultaneously,
the quencher pump was started and this activated the water
sprays in the quencher unit. At this time the cooling tower
fan was turned on, followed by the main blower which draws
the gas from the furnace through the F/C system. This com-
pleted the startup of the F/C scrubbing unit.
The operation of the F/C unit required little attention
and only routine visual inspection was necessary to ensure
that all equipment was performing as described. Fail-safe
circuits were provided so that the pumps would cease operation
in case of low water level in the liquor tanks. Similarly,
the blower would stop in case of liquor failure to the quencher.
During operation of the F/C scrubber unit, adjustments to
ensure adequate water rate and gas flow through the system
were made but were necessary only occasionally.
The shutdown procedure was simple. The main blower and
the cooling tower fan were turned off and all the pumps were
deactivated. This then stopped the operation of the scrubber
system.
PERFORMANCE TEST PROCEDURES
This section describes the test procedures that were used
to determine the performance of the F/C scrubber unit. The
particle and contaminant gas collection efficiency of the
system were measured by a combination of tests described below.
All tests were run after the scrubber system was in operation
and had achieved a steady state condition.
Modified EPA Method 5 (Filter Runs)
A modified EPA Method 5 test procedure was utilized to
determine overall particulate loading, particulate penetration
47
-------
and gas absorption in the system. The samples were obtained
at the furnace stack, quencher inlet, scrubber inlet, and scrub-
ber outlet.
Impaction Runs
The particle characteristics in the gas streams were mea-
sured by A.P.T. cascade impactors. The impactors were heated to
prevent vapor condensation on the impactor walls. Pre-cutters
were also used preceding the impactor in order to remove the
larger particles.
For both the filter and the impactor runs, the impinger
catch was analyzed to determine chloride ion concentration, pH
and total solids. The efficiency of the scrubber system for the
absorption of contaminant gases can be determined from these
data.
METHODS OF CALCULATION
The F/C scrubber system performance was measured by sam-
pling with both absolute filters and cascade impactors. The fil-
ter runs provided information on the total particulate loading,
and the overall scrubber efficiency. The impactor runs provided
particle characteristics such as the particle size distribution,
cumulative mass loading, particle penetration, and the particle
number concentration.
Particle Loading and Overall Efficiency
The" total particulate loading inlet and outlet to the system
and the overall efficiency were calculated as follows:
1. The dry sample volume flow rate was converter!, to normal
conditions at 0°C and 1 atm (DNm3/min).
2. Total weight gain on the sampling elements were combined
into one quantity (g).
3. The total particle mass loading, "C", (g/DNm3) was calculated
from:
c _ total weight gain (g)
sampling rate (DNm3/min) x sampling time (min)
(5-1)
48
-------
The overall penetration was calculated from:
C
Pi = -°-
C.
(5-2)
where
^ " and "GO " were the inlet and outlet particle
loadings measured simultaneously.
5. The overall efficiency (E%) was calculated from:
E* = (1-Ft) 100% (5-3)
Particle Size Distribution
The particle size distributions were measured gravimetrically
using the cascade impactor as a device to fractionate particles
of different size. Cumulative mass of particles collection on
each stage and all the stages below, including the absolute
filter were calculated as a percentage of the total weight gain.
The cut diameters for the impactor stages were calculated from
the sampling rate in conjunction with the calibration data for
A.P.T. impactors.
Cumulative Mass Loading and Particle Penetration
The penetration as a function of particle size, or the
fractional penetration through the system, was determined
by a stepwise graphical procedure. The procedure is based
on the following equation:
Pt
f(dpa)o _
*iVi
dC
ld(V.
' dC '
d(dpa).
outlet
inlet
(5-4)
where "dC/d(d )" is the slope of cumulative mass loading less
than "d " versus the aerodynamic particle diameter curve at
"dpa" and ec*uals "fCdpa)"' and "Pt Cdpa) is the penetration
for particle diameter "d ".
pa
To determine the penetration as a function of particle
size the following procedure was followed:
49
-------
Cumulative mass loading for all the stages and the
filter, below the stage with a cut diameter of "dpa",
were plotted against "d " from the inlet and
pa
outlet cascade impactor samples.
Slopes of the inlet and outlet plots above were deter-
mined for several McL_M values in the range of 0.5 ymA
pa
to 3.0 ymA. The fractional penetration was determined
for each "d " from the ratio of the slopes as described
pa c
above. The fractional penetration was then plotted over
the entire size range (Pt(d^ ) versus d „) to give the
pd pa
penetration grade curve.
50
-------
4. The overall penetration was calculated from:
C
Pt = -2- (5-2)
C.
where "(^ " and "GO " were the inlet and outlet particle
loadings measured simultaneously.
5. The overall efficiency (E%) was calculated from:
El = (1-Pt) 100% (5-3)
Particle Size Distribution
The particle size distributions were measured gravimetrically
using the cascade impactor as a device to fractionate particles
of different size. Cumulative mass of particles collection on
each stage and all the stages below, including the absolute
filter were calculated as a percentage of the total weight gain.
The cut diameters for the impactor stages were calculated from
the sampling rate in conjunction with the calibration data for
A.P.T. impactors.
Cumulative Mass Loading and Particle Penetration
The penetration as a function of particle size, or the
fractional penetration through the system, was determined
by a stepwise graphical procedure. The procedure is based
on the following equation:
dC
f(d__)_ dl
Pt
dC
outlet (5-4)
dCd-; inlet
where MdC/d(dpa)" is the slope of cumulative mass loading less
than "dpa" versus the aerodynamic particle diameter curve at
"dpa" and equals "f(dpa)", and "Pt (dpfl) is the penetration
for particle diameter "d "
pa
To determine the penetration as a function of particle
size the following procedure was followed:
49
-------
Cumulative mass loading for all the stages and the
filter, below the stage with a cut diameter of "
were plotted against "d " from the inlet and
pa
outlet cascade impactor samples.
Slopes of the inlet and outlet plots above were deter-
mined for several "d " values in the range of 0.5 ymA
pa
to 3.0 ymA. The fractional penetration was determined
for each "d " from the ratio of the slopes as described
pa
above. The fractional penetration was then plotted over
the entire size range (Pt(d _) versus d „) to give the
pd p 3.
penetration grade curve.
50
-------
CHAPTER 6
EXPERIMENTAL RESULTS AND DISCUSSION
After the construction of the F/C scrubber demonstration
plant, startup and performance testing were begun. Initial
startup runs were made with ambient air through the F/C scrubber
system. Once satisfactory performance of all mechanical equip-
ment had been achieved the hot flue gas from the furnace was
hooked up to the F/C scrubber unit. The system operated close
to the design conditions but higher particulate emission rate
was encountered at the source than previously monitored. Table
6-1 presents a brief summary of total particulate loadings at
the furnace stack for #1 grade copper wire at different stages
of the demonstration program. The emissions from the processing
of both #1 premium and #1 regular grade went up from those
measured before the initial site selection to the operation of
the demonstration plant.
The purpose of the testing phase of the program was to
obtain the overall mass efficiency of the system, the particle
collection efficiency across the F/C scrubber, the operational
\
reliability of the demonstration plant, and the economics of
F/C scrubbing. Thus, both performance and process data along
with operational information were determined and analyzed. The
measurement techniques and methods of data analysis and calcula-
tions of results are described in the preceding chapter. The
purpose of this section of the report is to present the results
and discuss the salient features of the test program. The eco-
nomics of the F/C scrubbing system are discussed in Chapter 8.
Overall, the F/C demonstration scrubber was capable of con-
trolling the emissions from the processing of #1 premium grade
copper wire. The performance was marginal for #1 regular grade
51
-------
TABLE 6-1. TOTAL PARTICULATE LOADINGS AT FURNACE
STACK DURING DIFFERENT STAGES OF F/C
SCRUBBING DEMONSTRATION PROGRAM.
Stage of
F/C Program
Site
Selection
Pilot
Plant
•
Demonstra-
tion F/C
Scrubber
Number of
Runs for
Averaging
2
4
4
20
Type of Wire
Processed
Premium #1
Regular #1
Premium #1
Regular #1
Average *
Loading
(g/DNm*)
0. 36**
1.3
0.78
1.8
* Mass loadings are based on overall average emissions
over charging cycle and include impinger catch.
** Corrected to 121 C02
52
-------
copper wire but the demonstration scrubber was not capable of
controlling the emissions from #2 and #3 grades of copper wire.
EXPERIMENTAL RESULTS
The F/C demonstration plant was evaluated at ten operating
modes as listed in Table 6-2. The purpose of the changes from
mode to mode was both to increase the efficiency of the F/C
system to cope with the higher emission rate than previously
monitored and to obtain collection efficiency data for
different modes of operation.
The performance of the system, based on mass concentration
data for different operating modes, is presented in Table 6-3.
This table gives the inlet and outlet concentrations of particu-
late matter, the corresponding penetrations, and the total
pressure drop across the system. The process data for these
runs are summarized in Table A-l of the Appendix "A".
Cascade impactor data were taken in order to determine
collection efficiency as a function of particle size. The
cumulative loading data are presented in Table 6-4, while
the corresponding impactor stage data are given in Appendix
"B". The cascade impactor data were utilized to calculate the
fractional penetrations as a function of particle aerodynamic
diameter. The calculation procedure was previously discussed
in Chapter 5. The resulting fractional penetrations for the
test along with the predicted penetration curves are presented
in Figures 6-1 to 6-18.
53
-------
TABLE 6-2.F/C SCRUBBER DEMONSTRATION PLANT
OPERATING CONFIGURATIONS
Conf. No
Description of System
5 plates in scrubber, 0.48 cm hole diameter, cold water
on pass A of scrubber. .
B
5 plates in scrubber, 0.48 cm hole diameter, cold water
on pass B of scrubber.
6 plates in scrubber, 4 plates with 0.48 cm hole dia-
meter, 2 plates with 0.32 cm hole diameter, cold water
on pass B of scrubber.
D
5 plates in scrubber, 3 plates with 0.48 cm hole dia-
meter, 2 plates with 0.32 cm hole diameter, cold water
on pass B of scrubber.
5 plates in scrubber, 3 plates with 0.48 cm hole dia-
meter, 2 plates with 0.32 cm hole diameter, cold water
on pass A of scrubber.
5 plates in scrubber, 3 plates with 0.48 cm hole dia-
meter, 2 plates with 0.32 cm hole diameter, cold water
on pass B of scrubber, sodium carbonate solution spray
in crossover duct, charge wetted with water.
Sampling at peak of emissions only, 5 plates in scrub-
ber, 3 plates with 0.48 cm hole diameter, 2 plates with
0.32 cm hole diameter, cold water on pass B of scrubber
sodium carbonate solution spray in crossover duct,
charge wetted with
H
Sampling at peak of emissions only, 5 plates in scrub-
ber, 3 plates with 0.48 cm hole diameter, 2 plates with
0.32 cm hole diameter, cold water on pass B of scrub-
ber, sodium carbonate solution spray in crossover duct,
charge wetted with water, water spray in crossover duct
Sampling at peak of emissions only, 6 plates in scrub-
ber, 4 plates with 0.48 cm hole diameter, 2 plates
with 0.32 cm hole diameter, cold water on pass B of
scrubber, sodium carbonate solution spray in crossover
duct, charge wetted with water, water spray in
crossover duct.
6 plates in scrubber, 4 plates with 0.48 cm hole dia-
meter, 2 plates with 0.32 cm hole diameter, cold water
on pass B of scrubber, sodium carbonate solution spray
in crossover duct, charge wetted with water, water
spray in cross over duct.
54
-------
TABLE 6-3.
F/C SCRUBBER DEMONSTRATION
PLANT SYSTEM PERFORMANCE
Run
No.
1
2
3*
4*
5
6
7*
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
Config-
uration
A
A
A
A
A
B
B
C
C
D
D
D
D
D
D
D
D
D
' D**
D**
E
E
APxlO'3
N/m2
4.2
4.2
4.3
4.2
4.3
4.2
4.2
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
' 5.0
5.0
5.0
5.0
5.0
5.0
5.0
Total Part. Conc.(g/DNm3)
qu.ln
1.3
1.6
1.7
2.1
1.2
5.7
1.2
2.6
1.1
Scr.ln
0.93
2.00
0.97
0.30
0.20
Scr.Out
0.24
0.35
0.29
0.35
0.37
0.11
0.057
0.18
0.076
0.37
0.23
0.39
0.099
0.30
0.28
0.22
0.060
0.096
0.73
0.22
0.18
0.090
Pt, Ou§Scr.
%
18.5
21.9
17.0
2.9
8.0
12.8
18.3
6.9
8.2
Pt, Scrub.
%
31.2
17.5
38.1
36.7
28.5
Qu.In = quencher inlet, Scr.ln = scrubber inlet,
Scr.Out = scrubber outlet
* Cascade impactor run ** Liquor to pass A only
Note: a Runs 1-56: Wire quality - number 1, regular grade
b Runs 57-78: Wire quality - number 1, premium grade
c Total particulate concentration includes both filter
and impinger catch.
55
-------
TABLE 6-3 (continued)
Run
No.
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42*
43*
44
_
Config-
uration
E
E
E
D
D
D
D
D
D
D
D
F
F
F
F
G
G
G
G
G
G
H
APxlO*3
N/m2
4.7
4.7
4.7
4.7
5.0
5.0
5.0
5.0
5.0
5.0
5.0
4.7
5.0
4.8
4.7
4.7
4.7
4.7
4.7
4.7
4.7
5.0
Total Part. Conc.(g/DNm3)
Qu.In
2.6
2.6
1.9
1.2
1.6
2.0
0.79
0.32
1.6
0.92
1.2
0.96
0.52
1.6
1.2
3.2
3.5
2.6
2.6
6.4
Scr .In
0.231
0.40
0.27
Scr. Out
0.66
0.83
0.41
0.20
0.12
0.22
0.13
0.098
0.31
0.22
0.36
0.090
0.096
0.13
0.18
0.21
0.054
0.14
0.12
0.11
0.29
Pt, Qu^Scr.
25.4
31.9
21.6
16.7
7.5
11.0
16.4
30.6
19.4
23.9
30.0
17.3
6.0
10.8
5.6
6.0
2.1
5.4
4.5
Pt, Scrub .
23.4
30.0
40.7
Qu.In = quencher inlet, Scr
Scr.Out = scrubber outlet
* Cascade impactor run
In = scrubber inlet
Note: a Runs 1-56: Wire quality - number 1, regular grade
Runs 57-78: Wire quality - number 1, premium grade
c Total particulate concentration includes both filter and
impinger catch.
56
-------
TABLE 6-3
(Continued)
Run
No.
45
46
47
48
49
50
51
52
53
54
55
56*
57*
58*
59*
Config-
uration
H
H
H
H
H
H
I
I
I
I
I
I
I
I
I
APxlO"3
N/m2
5.0
5.0
5.0
5.0
5.0
5.0
6.9
6.9
6.9
6.4
5.3
5.7
6.0
6.1
6.0
Total Part. Cone. (g/DNm3)
frurn .
Out
4.0
4.5
3.8
2.7
3.5
4.1
2.2
1.4
2.3
1.6
2.3
i^u.ln
1.3
0.68
0.83
1.4
bcr.Uut
0.22
0.27
0.15
0.17
0.24
0.21
0.12
0.10
0.094
0.19
0.20
0.077
0.023
0.030
0.031
Pt, Over-
all Syst.
% ft**
5.5
6.0
3.9
6.3
6.9
5.1
5.4
7.1
4.1
11.9
8.7
Pt, Qu^cr.
%
5.9
3.4
3.6
2.2
Furn.Out = furnace outlet, Qu.In = quencher inlet,
Scr.Out = scrubber outlet
* Cascade impactor run
*** Overall system penetration based on inlet mass concentration
measured at furnace out (upstream of sodium carbonate spray) and
outlet mass concentration measured at scrubber outlet.
o
Note: Runs 1-56: Wire quality - number 1, regular grade
Runs 57-78: Wire quality - number 1, premium grade
c Total particulate concentration includes both filter and
impinger catch.
Total particulate concentrations were measured at the
furnace outlet and not at the scrubber inlet for run 45
and those following it.
57
-------
TABLE 6-3,
(Continued)
Run
No.
60
61*
62*
63
64*
65
66*
67**
68
69*
70**
71*
72*
73*
74*
75*
76*
77*
78*
Config-
uration
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
APxlO"
N/m
6.1
5.8
6.1
6.0
6.0
5.8
5.8
6.1
5.8
6.0
6.1
6.0
6.1
6.8
6.8
6.3
6.1
6.5
6.5
Total Part. Conc.fe/DNm3)
Furn.
Out
1.4
0.53
0.67
0.53
Qu.In
0.89
1.1
2.8
1.2
1.9
1.0
1.6
1.3
1.3
1.2
1.5
2.5
1.2
1.2
Scr.Out
0.068
0.055
0.052
0.083
0.16
0.087
0.048
0.055
0.038
0.077
0.080
0.037
0.049
0.056
0.039
0.059
0.052
Pt, Over-
all Syst.
% ***
4.9
15.7
13.0
7.2
Pt, Qu§Scr.
%
6.2
4.7
5.7
4.0
2.9
7.7
6.2
2.8
4.1
3.7
1.6
4.9
4.3
Furn.Out = furnace outlet, Qu.In = quencher inlet,
Scr.Out = scrubber outlet
* Cascade impactor run
** Blank run
*** Overall system penetration based on inlet mass concentration
measured at furnace out (upstream of sodium carbonate spray)
and outlet mass concentration measured at scrubber outlet.
Note: a Runs 1-56: Wire quality - number 1, regular grade
Runs 57-78: Wire quality - number 1, premium grade
c Total particulate concentration includes both filter and
impinger catch.
Total particulate concentrations were measured at the
furnace outlet and not at the scrubber inlet for run 45
and those following it.
58
-------
TABLE 6-4. PARTICULATE AND IMPINGER DATA*, F/C SCRUBBER DEMONSTRATION PLANT
Run
No.
3
4**
7**
42**
43**
56
58
59
61
62
64
66
69
71
72
73
74
75
76
77
78
Total Particulate Cone.
In(g/DNm3)
1.7
2.0
0.20
0.40
0.27
1.3
0.83
1.4
0.89
1.1
2.8
1.2
1.0
1.6
1.3
1.3
1.2
1.5
2.5
1.2
1.2
Out(g/DNm3)
0.29
0.35
0.057
0.12
0.11
0.077
0.030
0.031
0.055
0.052
0.16
0.048
0.077
—
0.081
0.037
0.049
0.056
0.039
0.059
0.052
Impactor Part. Cone.
In(g/DNm3)
0.089
1.9
0.11
0.39
0.27
1.2
0.83
1.3
0.89
1.1
2.5
1.2
0.70
1.2
0.97
1.1
1.1
1.2
2.2
0.76
1.2
Out(g/DNm3)
0.024
0.34
0.034
0.12
0.11
0.048
0.030
0.031
0.053
0.052
0.14
0.048
0.077
0.081
0.037
0.049
0.048
0.036
0.051
0.014
Impt.Pt
1
27.0
17.9
30.9
30.8
40.7
4.0
3.6
2.4
5.9
4.7
5.6
4.0
11.0
8.3
3.4
4.4
4.0
1.6
6.7
1.2
Impinger Cone.
In(g/DNm3)
1.6
0.09
0.09
0.01
0.0
0.09
0.0
0.12
0.0
0.28
0.01
0.32
0.36
0.36
0.19
0.034
0.27
0.27
0.43
0.02
Out(g/DNm3)
0.26
0.01
0.02
0.0
0.0
0.03
0.0
0.0
0.0
0.02
0.0
0.0
0.0
0.0
0.0
0.01
0.0
0.01
0.04
«
in
* Particulate data taken across quencher inlet and scrubber outlet unless other-
wise noted.
** Particulate data taken across plate scrubber only.
Note: a Run 7 was a cold impactor run.
b Inlet data for run 71 (outlet) was inadequate due to error in test
procedure.
f
Outlet loadings for run 78 were too light to provide adequate
size data.
-------
1.0
EXPERIMENTAL
PREDICTED —
0.05
0.3 0.5
1.0 1.5 2.0 3.0
dpa,
Figure 6-1. Particle penetration versus aero-
dynamic diameter for run 3.
1.0
o
I-H
H
U
2
O
E-i
m
2
W
w
a o.i
I EXPERIMENTAL
PREDICTED —
0.2
0.05
0.3
0.5
1.0 1.5 2.0 3.0
dpa>
Figure 6-2. Particle penetration versus aero-
dynamic diameter for run 4.
-------
1.0
§
H 0.5
u
PL,
2
O
OH
H
w
m
i-j
u
i— i
E-
0.4
0.3
0.2
0.1
0.05
EXPERIMENTAL
PREDICTED
0.3 0.5
1.0 1.5 2.0 3.0
dpa>
Figure 6-3. Particle penetration versus aero-
dynamic diameter for run 42.
1.0
o
U
0.5
0.4
0.3
2 0.2
H
m
u
hH
H
fX,
0.1
0.05
pa>
Figure 6-4. Particle penetration versus aero-
dynamic diameter (scrubber only)
for run 43.
Note: Predicted penetration curve was very
low due to very low predicted particle
number concentration for this run.
(See Table E-l of appendix)
-------
0.3
c-o
EXPERIMENTAL
PREDICTED
0.01
0.3
0.5
1.0
1.5 2.0 3.0
dpa,
Figure 6-5. Particle penetration versus aero
dynamic diameter for run 56.
z
o
o
i—i
H
g
H
PH
W
0.5
0.4
0.3
0.2
0.1
0.05
0.04
0.03
0.02
EXPERIMENTAL
PREDICTED —
0.3 0.5
1.0 1.5 2.0 3.0
V' ymA
Figure 6-6. Particle penetration versus aero-
dynamic diameter for run 58.
-------
§
I—I
E-H
U
A
2
O
%
E-
m
2
w
PH
M
J
U
I—I
E-H
rt
<
PH
EXPERIMENTAL
-•aPREDICTED
0.02
0.3 0.5
1.0 1.5 2.0 3.0
d , ymA
pa.
Figure 6-7. Particle penetration versus aero-
dynamic diameter for run 59.
1.0
EXPERIMENTAL
PREDICTED —
0.05
0.3 0.5
1.0 1.5 2.0 3.0
dpa'
Figure 6-8. Particle penetration versus aero-
dynamic diameter for run 61.
-------
EXPERIMENTAL
PREDICTED
0.02
0.3 0.5
1.0 1.5 2.0 3.0
dpa,
Figure 6-9. Particle penetration versus aero-
dynamic diameter for run 62.
EXPERIMENTAL
PREDICTED
0.3 0.5 1.0 1.5 2.0 3.0
d , umA
pa' K
Figure 6-10. Particle penetration versus aero-
dynamic diameter for run 64.
-------
0.5
EXPERIMENTAL
PREDICTED
1.0 1.5 2.0 3.0
0.02
Figure 6-15. Particle penetration versus aero-
dynamic diameter for run 74.
o
H
W
Z
m
U
d,
EXPERIMENTAL
PREDICTED —
0.3
0.5
1.0 1.5 2.0
pa>
3.0
Figure 6-16. Particle penetration versus aero-
dynamic diameter for run 75.
-------
0.7
ON
OO
EXPERIMENTAL
PREDICTED —
0.02
0.3
1.0
dpa, pmA
1.5 2.0 3.0
Figure 6-17. Particle penetration versus aero-
dynamic diameter for run 76.
0.7
g
i—i
H
U
o
I—I
H
H
W
W
H
05
<
O,
EXPERIMENTAL
PREDICTED —
0.03
0.3
0.5
1.0 1.5 2.0 3.0
dpa,
Figure 6-18. Particle penetration versus aero-
dynamic diameter for run 77.
-------
DISCUSSION OF RESULTS
Operating Modes
Operating modes and scrubber configuration were varied in
order to cope with the high emission rates encountered at the
source. This provided a base for comparison of performance
under different sets of process and equipment parameters. In
summary, the following variations were studied.
1. The F/C scrubber system was provided with the provision
of adding a sixth plate inside the unit. The purpose of adding
the extra plate was to enhance the efficiency of the scrubber
through increasing collection of particles by inertial impaction.
This in turn resulted in higher pressure drop across the system.
2. The F/C scrubber was modified by installation of sieve
plates with perforation of smaller diameter at two positions in-
side the scrubber. The existing plates had a 0.48 cm (3/16")
diameter perforation with 6.0% open area, while the modified plate*
used 0.32 cm (1/8") perforation with 3.3% open area. This change
achieved a greater velocity through the perforation, resulting
in higher collection efficiency due to inertial impaction.
3. The sieve plate scrubber consisted of two passes of ver-
tical columns connected in series. The gas flowed into the
bottom of pass 'A1 equipped with three sieve plates, then into
the bottom of pass 'B1, which also was provided with three sieve
plates. , The piping system was designed with the flexibility of
injecting the cold water from the cooling tower at either pass
of the scrubber, with the other pass receiving the hot water
from the scrubber sump tank. The system was operated with cold
water injected at pass 'A' during one operational mode and simi-
larly with cold water injected at pass 'B1 during another opera-
ting mode. It was observed, that for this particular design
of the F/C scrubber, introduction of cold liquor at pass 'B' of
the scrubber versus hot liquor did reduce emission of excessive
steam from the stack of the F/C unit.
4. The emissions from the furnace were cyclic in nature and
were higher during the initial part of the charging cycle. Emis-
69
-------
sion data were collected on both an overall average basis and
also on a peak interval time basis only. The peak interval
testing provided the emissions data during the worst part of
the charging cycle.
5. The scrap charge was also pre-wetted during the later
operating modes before introduction into the furance. The
purpose was to increase the gas humidity of the flue gas and
consequently provide a higher condensation ratio across the F/C
unit.
6. Initially the F/C demonstration unit was built with a
carbon steel cross-over duct to convey the pollutant gas from
the furnace stack to the F/C unit. After corrosion caused its
failure, this cross-over duct was replaced with a stainless steel
(316) duct. A sodium carbonate spray solution was added inside
the cross-over duct for the purpose of reducing the formation of
metal chloride particles and to neutralize the acidity of the
gas stream at an early stage. Water sprays were also added in
the same duct to achieve further improvement in performance.
Mass Loading Versus Wire Scrap Quality
The performance results of the F/C unit under different
operating models are given in Table 6-3. The performance results
indicate that for a majority of cases, the system operated at an
efficiency of over 90%. As regards the mass loading, two grades
of #1 wire were processed at the host facility. The two grades
were regular #1, which is 601 copper and 40% plastic, and premium
#1, which is 80% copper and 20% fabric (with some rubber). This
change in wire quality was reflected in the total particulate loadings
at both the inlet and outlet to the F/C system. Overall, the *
performance indicated that the system was well within the local
air pollution control district standards during the reclamation
of premium #1 copper wire. The performance of the system during
the reclamation of #1 copper wire was marginal due to limitations
on the fan pressure capacity.
Performance Model
The predictability of the F/C demonstration plant can be
70
-------
2
5
Cn m
»J
u
0.7
0.5
0.4
0.3
w 0.1
0.05
0.04
0.03
0.02
0.3 0.5
RUN 66
EXPERIMENTAL
PREDICTED
1.0 1.5 2.0 3.0
dpa,
Figure 6-11. Particle penetration versus aero-
dynamic diameter for run 66.
z
o
hH
H
U
o
i—i
H
m
m
a,
w
j
u
i—i
H
01
0.5
0.4
0.3
0.2
0.1
0.05
0.04
0.03
0.02
RUN 69
EXPERIMENTAL
PREDICTED
0.3 0.5
1.0 1.5 2.0 3.0
dpa, ymA
Figure 6-12. Particle penetration versus aero-
dynamic diameter for run 69.
-------
ON
o
CJ
O
m
U
i — i
H
Di
0-.5
0.4
0.3
0.2
0.1
0.05
0.04
0.03
0.02
EXPERIMENTAL
PREDICTED —
0.3 0.5
1.0 1.5 2.0 3.0
dpa' WniA
Figure 6-13. Particle penetration versus aero-
dynamic diameter for run 72.
PU
z
o
H
ra
z
m
cu
0.4
0.3
0.2
0.1
2 0.05
0.04
0.03
0.02
0.3
0.5
1.0 1.5 2.0
3.0
dpa, urn
Figure 6-14. Particle penetration versus aero-
dynamic diameter for run 73.
Note: Predicted penetration curve was very
high due to very high predicted particle
number concentration for this run.
(see Table E-l of appendix)
-------
judged from a comparison of the actual performance of the system
with the predicted performance as determined from the design
model. The details on the design model are given in the next
chapter. This section compares the results of the F/C plant
with the predicted performance.
The observed performance was close to the predicted
performance as calculated using the design model. Figures 6-1
to 6-18 give a comparison of experimental particle penetration
of the F/C demonstration scrubber to the predicted penetration.
Table 6-5 summarized the results, showing the experimental
overall penetration of the demonstration scrubber versus the
calculated overall penetration. The procedure for predicting
penetration is illustrated by an example in Appendix "F".
As indicated by the data, there is some deviation between
the experimental and predicted performance. Some of the reasons
for this deviation are given in the following list of assump-
tions and uncertainties in the model and the experimental data.
1. The calculation procedure does not account for collec-
tion in the quencher section of the system. The quencher collects
the larger particles, but not the submicron particles.
2. The flux force mechanism of thermophoresis and the
mechanism of centrifugation were not considered in the calcu-
lation of the performance using the model. It was assumed
that the effect of inertial impaction and diffusiophoresis
would overshadow the effect of other mechanisms.
3. The foam density was considered to be of constant
value (F=0.4) in the design model prediction. In actuality,
there would be some variation due to changes in plate
hydrodynamics under different operating conditions.
4. The design model assumes the particle to be
wettable but insoluble. Since the particles were quite
soluble in water they will cause more condensation than
predicted and the performance for the system should be
better than predicted.
5. The fraction of water vapor condensing on the particles
71
-------
TABLE 6-5. COMPARISON OF EXPERIMENTAL DATA
VERSUS PREDICTED PENETRATIONS FOR
F/C SCRUBBER DEMONSTRATION
Run
No.
3
4**
42**
43**
56
58
59
61
62
64
66
69
72
73
74
75
76
77
Experimental *
Loading (mg/DNm )
Inlet
89.0
1,910.0
395.0
274.0
1,170.0
829.0
1,250.0
886.0
1,150.0
2,490. 0
1,230.0
701.0
969.0
1,100.0
1,140.0
1,200.0
2,240.0
763.0
Outlet
23.9
338.0
121.0
106.0
48.4
30.2
31.2
53.3
51.7
140.0
48.2
77.1
80.7
36.7
49.0
48.3
36.0
50.5
Penetration
1
26.8
17.7
30.6
38.7
4.1
3.6
2.4
6.0
4.5
5.6
3.9
11.0
8.3
3.3
4.3
4.0
1.6
6.6
Predicted
Penetration
%
27.7
18.3
14.2
***
3.7
1.3
4.1
2.6
4.5
5.1
8.6
4.7
8.0
****
2.6
6.0
2.7
4.4
* Particulate data across quencher inlet and scrubber outlet
unless otherwise indicated.
** Particulate data across plate scrubber only.
*** Predicted particle number concentration was very low.
**** Predicted particle number concentration was very high.
72
-------
was approximated to be 0.25 of the total condensation in the
scrubber in the design model computations. This approximation
was based on simulated computer runs using a model for F/C
scrubbing in a sieve plate at conditions comparable to the F/C
process parameters at the demonstration plant. This value will
vary as indicated in Chapter 7, depending on different process
conditions such as the gas-phase temperature, liquid-bulk
temperature, number-particle concentration, and liquid-phase
heat transfer coefficient.
6. The design model uses as one of its parameters the
particle number concentration at the inlet to the F/C system.
As discussed in Appendix "E". there is uncertainty in the
procedure used to determine number concentration. Any error
in these predictions causes deviations between the predicted
and actual performance. An attempt is made in Figure 6-19 to
to study the effect of the absolute values of the predicted
particle number concentration "n" on the agreement between
predicted particle penetration and experimental particle
penetration. This plot shows that the agreement is best
/ Predicted Pt -. „ \ , ,
( Ex^erTielnIT-pT = 1'°) when the Particle no. concen-
tration is about 1.5 x 108 /DNcm3.
, The ratio of penetrations is higher when the predicted
particle no. concentration is higher and lower when the
predicted particle no. concentration was lower. There is
uncertainty and great variation in the absolute values of
the particle number concentration. Therefore, the graph
indicates that some of the deviations of the model predictions
from the experimental prediction could very well be due to
the deviations of the predicted particle number concentrations
from their true values.
Error Analysis
The accuracy of the experimental results, as discussed
in this chapter, is dependent on the independent and cumu-
lative errors of several measurements. The scope of this
73
-------
2.5
o
i—i
H
2.0
1.5
1.0
0.5
3x10
5xl07
IxlO8 2xl08
PARTICLE NO. CONCENTRATION, NO/DNcm3
5xl08
1x10 9
Figure 6-19
Effect of calculated particle number concentration on
agreement between predicted and experimental
for 0.5 ymA diameter particles.
-------
report does not permit a complete statistical error analysis
on all the experimental results. However, the individual
errors associated with total filter runs and impactor size
distributions are discussed below.
The independent variables involved in the determination,
along with their precision and nominal values, are listed in
Table 6-6 and 6-7. The equation for calculating the overall
penetration and size distribution involves both intermediate
variables and multi-appearances of variables. However, if
these are replaced with an exact expression, the fractional
error in the dependent variable "C" can be calculated as:
dc - 1 9C I Jv 1 3C I ar I
C- - C 3*! I n. dX> + -C 3X2 I „. d*' + • • -fj | n 3X (6-1)
n
where xi5 X2, ..-X = independent variables as listed in
Tables 6-6 and 6-7.
C = absolute value of dependent variable
calculated at nominal conditions
3X}, 3Xa, . . . 3X = precision of independent variable.
J3 (~*
~— = fractional error of dependent variable
Here the quantity
si = r IY T, (i = 1, 2, ...n.) (6-2)
LI o A i n.
represents the sensitivity of error ^— to an error 3X^. This
value gives a direct indication of the effect of an error
in the measured variable on the fractional error of the
dependent variable.
The errors involved in the calculation of fractional
penetration and inlet number concentration are more subjective
in nature for they involve judgment in analyzing graphs and
determining slopes.
75
-------
TABLE 6-6. MEASURED VARIABLES, DEFINITION, PRECISION AND TEST
CASE NOMINAL VALUES FOR TOTAL EFFICIENCY CALCULATION
Symbol
VDGM
t
mg
Ps
Pb
Measurement
Volume of dry gas
read by dry gas
meter
Length of time of
test
Sample weight
Orifice pressure
reading dry gas
meter
Barometric
pressure
Unit
ft3
min
mg
in H20
in Hg
Precision
0.002 .
.2/60
0.50
0.02
0.10
Nominal
Value
10.0
60.0
50.0
0.15
29.9
TABLE 6-7. MEASURED VARIABLES, DEFINITION, PRECISION AND TEST
CASE NOMINAL VALUES FOR SIZE DISTRIBUTION CALCULATION
Symbol
VDGM
t
mg
Ps
pb
TA
V
Ws
Measurement
Volume of dry gas
read by dry gas
meter
Length of time of
test
Sample weight on
each impactor
substrate
Orifice pressure
reading dry gas
meter
Barometric pressure
Impactor temperature
Volume of impinger
Weight of dryer
Unit
ft3
min
mg
in H20
in Hg
°C
ml
g
Precision
0.002
.2/60
0.02
0.05
0.10
1.0
0.5
0.01
Nominal
Value
2.5
5.0
0.5
0.15
29.9
50.0
200.0
200.0
Note: Tables 6-6 and 6-7 include English units where test
apparatus was equipped accordingly.
76
-------
CHAPTER 7
F/C SCRUBBER PERFORMANCE PREDICTION MODEL
The purpose of the F/C scrubber performance model is to
predict the overall efficiency of a flux force condensation
scrubber under different sets of operating and process condi-
tions. The model can be used to predict the outlet concen-
tration of particulate matter that would be anticipated from
a F/C scrubber installation on a pollutant source with a known
particle concentration and size distribution of emissions. In
summary the F/C model first predicts the size distribution of
the grown particle and then determines the fractional penetra-
tions for different particle sizes. Using these penetration
data and the initial inlet size distribution, one can then
calculate the anticipated overall efficiency.
Under previous EPA contracts, Air Pollution Technology, Inc
conducted detailed studies on the technical and economic
feasibilities of applying F/C scrubbing for fine particle collec-
tion. These studies (Calvert et al. (1973 and 1975)) included
theoretical development of design equations for F/C scrubbing
and the development of a computer mathematical model to predict
particle collection in a sieve plate F/C scrubber. The predic-
tions compared well with experimetal results, but the compu-
terized model had certain limitations. Its usefulness for quick
and overall design calculations was limited since solutions for
each sieve plate and for each different set of operating condi-
tions had to be calculated individually. In this study, a revised
mathematical model was developed from the same basic relationships
and the design procedures were greatly simplified for prediction
of F/C scrubber performance.
77
-------
In the course of refining our design method we arrived at
the conclusion that the flux force effects would be treated
separately from the condensation effects in many scrubber situa-
tions. While the condensation-induced improvement in inertial
impaction efficiency could be handled conveniently, the flux
force deposition prediction remained cumbersome. Recently we
became aware of a concept developed by Whitmore (1977) which
provided the key to simplifying the prediction of the flux
force effects in F/C scrubbing. By incorporating Whitmorefs
concept into our model we have developed a much simpler design
method which is convenient to use. This revised model will be
described below.
BASIC CONCEPTS
Before proceeding to the details of the mathematical model
the basic concepts and outline of the approach will be dis-
cussed. If we consider a typical F/C scrubbing system, it
might have the features shown in Figure 7-1. The gas leaving
the source is hot and has a water vapor content which depends
on the source process. The first step is to saturate the gas
by quenching it with water. This will cause no condensation
if the particles are insoluble, but will if they are soluble.
There will be a diffusiophoretic force directed away from the
liquid surface.
Condensation is required in order to have diffusiophoretic
deposition, any growth on insoluble particles, and extensive
growth on soluble particles. Contacting with cold water or a
cold solid surface is employed to cause condensation. While
condensation occurs there will be diffusiophoretic and thermo-
phoretic deposition as well as some inertial impaction (and,
perhaps, Brownian diffusion). The particles in the gas leaving
the condenser will have grown in mass due to the layer of water
they carry.
Subsequent scrubbing of the gas will result in more par-
ticle collection by inertial impaction. This will be more
efficient than impaction before particle growth because of the
-------
WATER
WATER
WATER
CLEAN GAS
55°C
0.12g/g
IMPACTOR
o SAT .A GAS
55 C *"*
0.12g/g
CONDENSER
SAT.AGAS
74°C **
0.36g/g
SATURATOR
HOT AGAS
1,000°C T
O.Olg/g I
Figure 7-1. Generalized F/C Scrubber System.
79
-------
greater inertia of the particles. There may be additional
condensation, depending on water and gas temperatures, and
its effects can be accounted for as discussed above.
One can apply this general outline of F/C scrubbing to a
variety of scrubber types and Figure 7-2 shows a multi-plate
F/C scrubber system. It can be seen that the gas is satu-
rated before entering the plate column, although this is not
always necessary. The first plate can serve as the saturator
and partial condenser. Generally, the efficiency of heat and
mass transfer is so high (say, 80%+) on a well designed plate
that most of the condensation occurs on the first plate.
In subsequent plates the gas is scrubbed by inertial
impaction and there will be a minor amount of additional
condensation. We have shown a simple counter-current column
but other variations are possible.
The mathematical model is based on the process just des-
cribed for a plate-type F/C scrubber. It is outlined below:
I. Saturate the gas before it reaches plate 1
A. Particles are collected at size "d ".
B. No condensation occurs on the particles
(they are assumed insoluble).
II. Contact on plate 1
A. Particles are collected by impaction in the
bubble formation zone, still at size "d ".
Pi
B. Condensation occurs and particles grow to
"V-
C. Diffusiophoretic deposition removes some
particles from the gas in the froth layer on
the plate. Thermophoresis and centrifugal
deposition are neglected.
III. Contact on plate 2
A. Particles are collected by impaction in the
bubble formation zone at size "d ".
pa
B. Negligible condensation occurs.
IV. Contact on subsequent plates has same characteristics
as plate 2.
80
-------
CLEAN GAS
GAS
COLD
WATER
WATER
4
SAT.
WATER
_| Plate 2
3FB~";
•*-«u>v-'»»|vww'»*
Plale 1
WATER
Figure 7-2. Multiple plate F/C scrubber system.
-------
Although the model is somewhat idealized, it is well within
the bounds of engineering accuracy and the precision of experi-
mental measurements. One can always revise any stage of the
model with a closer approximation of parameters if one wishes
to examine the sensitivity of the method to parametric values.
In our evaluations of the model we found that further refinements
did not produce significant changes in predictions.
DIFFUSIOPHORETIC DEPOSITION
Particle deposition by diffusiophoresis was described by
the following equation in our previous models:
cm/s
pD [y/Ml + (i-y)/M2 (i-y) \dr,
or,
where D^ = diffusivity of water vapor in carrier gas, cm2/s
MI = molecular weight of water, g/mol
M2 = molecular weight of non-transferring gas, g/mol
y = mole fraction water vapor, dimensionless
r = distance in the direction of diffusion, cm
The molecular weight and composition function represented
by "Ci" describes the effect of molecular weight gradient on the
deposition velocity corresponding to the net motion of the gas
due to diffusion (the "sweep velocity"). For water mole frac-
tion in air ranging from 0.1 to 0.5, "Ci" varies from 0.8 to 0.8!
We used a rough average of 0.85 for "Ci" for computing "U p."
and consequent particle collection efficiency by integrating
over the period of condensation.
Whitmore concludes that the fraction of particles removed
from the gas by diffusiophoresis is equal to either the mass
fraction or the mole fraction condensing, depending on what
theory is used for deposition velocity. In other words, it
-------
is not necessary to follow the detailed course of the conden-
sation process, computing instantaneous values of deposition
velocity, and integrating over the entire time to compute the
fraction of particles collected. One can simply observe that
if some fraction of the gas is transferred to the liquid phase
it will carry along its load of suspended particles.
We have used Whitmore's general concept but with two modifi-
cations. First, one can see from equation (7-1) that Whitmore's
theory would be comparable to assuming that the particles move
with the same velocity as the gas phase. We have chosen to
retain the correction for molecular weight gradient, which
means that we will compute the particle collection efficiency
as 85% of the volume fraction of gas condensing on the cold sur-
face.
The second modification concerns how to compute the proper
value of the volume fraction of gas condensing. The problem is
that not all of the condensate goes to the heat transfer surface;
some of it goes to the suspended particles. As will be show^n in
detail later, the fraction of the condensate which causes par-
ticle growth depends on several factors and ranged from about
0.1 to 0.4 of the total condensate for the range of parameters
we explored.
If one is concerned only with diffusiophoretic deposition
the particle collection efficiency would, therefore, be 60% to
801 of that computed without accounting for condensation on
particles. In the case of a scrubber which also employs inertial
impaction the particles would be agglomerated to some extent by
the diffusional sweep, so they would have higher mass and be
easier to collect.
Without going into a detailed model of this phenomenon one
could use either of two simplifying assumptions:
1. Assume that the condensation on particles causes no
agglomeration.
2. Assume that the condensation on particles causes
agglomeration and that the inertial impaction
efficiency is sufficiently high that all of the
83
-------
particles swept to other particles will be collected
by impaction.
The first assumption will lead to too low an efficiency and the
second to too high an efficiency. However, the maximum differ-
ence between the two for a representative case of 25% of the
volume condensing and 25% of that going to the particles would
be 5.31. That is, the percentage of particles which could be
swept to other particles = (0.25 x 0.85 x 0.25 x 100). This is
a relatively small effect compared to the other uncertainties.
PARTICLE GROWTH
Particle growth is dependent on how well the particles can
compete with the cold surface for the condensing water. There
are several transport processes at work simultaneously in the
condenser section of an F/C scrubber:
1. Heat transfer
a. From the gas to the cold surface
b. From the particles to the gas
2. Mass transfer
•
a. From the gas to the cold surface
b. From the gas to the particles
A mathematical model which accounted for these transport
processes in addition to particle depostion has been described
in EPA reports, see Calvert et al. (1973 and 1975). The portions
of that model relating to particle deposition were deleted to
provide a model which would describe particle growth in the
absence of deposition. The basic relationships involved are as
follow:
The rate of change of particle radius is given by a mass
balance,
— ^i- , cm/s ^'~J)
where:
2 DG P
k'pG = RT—cT~15— = Particle to Sas mass transfer coeffi-
1 G p PBM cient, gmol/cm2-s-atm
(7-4)
84
-------
3
p • = water vapor partial pressure at vapor-particle inter-
p = molal desntiy of water, gmol/cm3
pG = water vapor partial pressure in bulk of gas
bubble, atm
PBM = mean partial pressure to non-transferring gas, atm
r = particle radius, cm
TV = gas bulk temperature, °K
PM = molar density of water, gmol/cm
face, atm
molal des
P = total pressure, atm
R = gas law constant, atm-cm3/gmol-°K
d = particle diameter, cm
Particle temperature can be computed from an energy balance
hr,r (T^-Tr) * /PP CPP M d Tpi = k' r LM (Pr-p^-) (7-5)
pb pi b I—*- &*• — I • , . £— pb Mb rpl
where:
2k
h r = T— = particle to gas heat transfer coefficient, (7-6)
T3v3 Ct -i/P OT/-
^ p cal/cm -s- K
where C = heat capacity of particle, cal/g-°K
k = thermal conductivity of gas, cal/cm2-s-°K/cm
LM = latent heat of vaporization for water, cal/gmol
t = time, s
T . = temperature at particle interface, °K
p = particle density, g/cm
The overall energy balance for the gas-liquid interface is
given by:
k'G at LM (PG
hT a. (TT. -
L t LI
- PLi) Ap dZ =
T 1 A HZ + h
1 -r 1 /A. U- Ll * H/-»
L p b
Ap dZ
85
-------
where k'G = mass transfer coefficient, gas to liquid,
gmol/cm2-s-atm
a = interfacial area for transfer
volume of scrubber, cm2/cm3
A = cross-sectional area of scrubber, cm2
P
hr = heat transfer coefficient, gas to liquid,
b cal/cm2-s-°K
T, = temperature of liquid bulk, °K
P . = water vapor partial pressure at vapor-liquid
interface, atm
dz = height of bubble rise, cm, in time "At", s
TT . = temperature at liquid-gas interface, °K
L1
hy = liquid phase heat transfer coefficient, cal/s-cm -°K
The equations given above are used along with enthalpy and
material balances for the total system of gas, liquid, and sus-
pended particles to form a mathematical model for condensation
and growth. The model was solved through a finite difference
method on an electronic computer for several situations which
are discussed below.
PREDICTION OF CONDENSATION
The condensation model was used to predict the fraction
of the total condensate which goes to the particles (this
fraction defined as "fn") as a function of several parameters.
The conditions investigated are as follows:
\
1. Scrubber type - one sieve plate
2. Inlet gas - saturated from 310°K to 350°K
3. Water - uniform bulk temperature from 310°K to 325°K
4. Particle number concentration - 107 to in9 /cm3
5. Particle diameter - 0.1 to 1.0 urn
6. Liquid phase heat transfer coefficient - 0.01 to 0.1
cal/cm2-s-°K
7. Condensation can occur when the gas is saturated
The computed values are plotted on Figures 7-3 through 7-7.
As can be seen, the figures show the following:
86
-------
00
c
o
o
a
M
1.0
0.8
o 0.6
i
§
in
z
HJ
0.4
O
u
1
o,
hL • O.lcal/s -csi2-°K
0.4 0.6
PARTICLE DIAMETER (d )
1,0 1,1
Figure 7-3. Predicted particle condensation ratio (f )
as a function of particle diameter. P
0.5
0.10 cal/s cm2-"K
30.S 310 315 320 325 330
LIQUID BULK TEMPERATURE (TL), °K
Figure 7-4. Predicted particle condensation ratio
Cfpl as a function of liquid bulk
temperature.
-------
1.0
-------
1.0
H
U
O
C/D
2
W
O
2
O
U
H
rt
<
PL,
0.8 ::::!
0.2
0.01
0.02
0.04
0.06
0.08
0.10
LIQUID PHASE HEAT TRANSFER COEFFICIENT, (hT),
cal/s cm2-°K L
Figure 7-7. Predicted particle condensation ratio (fp)
as a function of liquid phase heat transfer
coefficient.
89
-------
Figure 7-3. "f " does not depend much on "d "
7-4. "f " decreases significantly with "TL"
7-5. "f " decreases with "T " to an extent which
p b
depends on MTL"
7-6. "f " increases slightly with "n " above 107/cm3
7-7. "f " increases with "hL" up to "hL"
* 0.1 cal/s-cm2-°K
It was found in other computations that "f " decreases
significantly with "n " below about 106 particles/cm3. Since
industrial emissions generally have particle number concentra-
tions on the order of 107 and greater "n " has no significant
effect on "f " in practice. The liquid phase heat transfer
coefficient is an important parameter but, unfortunately, predic-
tions of its magnitude vary considerably depending on which
correlation is used. The value of 0.1 appears to be the best
supported by the literature for mass transfer on perforated
plates.
For a combination of parameters such as might be encountered
in a practical situation a value of "f " * 0.25 appears to be
reasonable. Given this one can compute the amount of particle
growth that will result from a given condensation ratio (i.e.,
g water condensed/g dry gas = q1 = condensation ratio). If the
particle size distribution and the scrubber characteristics are
known one can predict the overall penetration that will be
achieved in the scrubber.
INTERIAL IMPACTION DURING BUBBLE FORMATION
During the formation of bubbles on a sieve plate the jets
of gas emerging from the perforations impact on the liquid.
Particles are thus deposited on the liquid surface by inertial
impaction. Particle collection can be determined from:
40
Ptj = exp
90
-------
where F = foam density, volume fraction liquid
d = particle diameter, cm
p = particle density, g/cm3
Cf = Cunningham slip correction factor, dimensionless
u^ = gas velocity in the perforation, cm/s
VU = gas viscosity, poise
d, = diameter of perforation, cm
Pt• = penetration of particles of diameter, d , fraction
PERFORMANCE PREDICTION METHOD
The sequence of steps to be followed in predicting the perfor-
mance of a F/C scrubber system involving a sieve plate column
is as follow:
1. Determine the initial particle size distribution.
2. Compute particle penetration from the saturator (PtJ based
d
on the saturator collection efficiency characteristics and the
initial particle size distribution. No growth occurs in the
saturator.
3. Compute particle penetration due to inertial impaction
during bubble formation on the first plate (Pt^)• Use the particle
size distribution leaving the saturator and the collection
efficiency relationship for sieve plate given in equation (7-8).
4. Calculate the condensation ratio corresponding to the
scrubber operating conditions, from this compute "fv", the
volume fraction of gas condensing, and then calculate the pene-
tration due to diffsiophoresis (Ptc) according to equation (7-9) for
a conservative estimate or equation (7-10) for an optimistic
estimate
Ptc = 1 - 0.85 (fy) (1 - fp) (7-9)
Ptc = 1 - 0.85 £y (7-10)
where:
f _ moles HzO condensed _ q'
v "total moles originally in vapor Hi + 18.
29
91
-------
where :
HI = original humidity ratio, g/g
The diffusiophoretic penetration applies equally to all
particle sizes so it will not change the size distribution but
will decrease the particle concentration.
5. Determine the particle size distribution leaving the
condenser from the values of "q"' and "f "•
6. Compute the particle penetration function for the
remaining stages of the scrubber, based on the penetration for
one stage given by equation (7-8). The penetration for a given
particle diameter on one stage is "Ft-". For "N" stages of
equal efficiency the penetration for a given particle diameter
is "Pt-N".
1 N"
7. Use the relationship between "Pt. and "d " from step 5
and the grown size distribution from step 5 to compute the
overall penetration due to inertial impaction after growth (P^) •
8. To summarize, the total overall fractional penetration
for the F/C scrubber Pt will be the product of the following:
a. "Pt " due impaction in the saturator
b. "Pt" ^ue to imPacti°n in tne condenser
c. "Pt " due to diffusiophoresis in the condenser
d. "Pty* due to impaction in stages after the condenser.
Thus ,
Ptt = Pta x Ptb x Ptc x Ptd (7-11)
In order to determine the overall penetration of the system
PT, the penetration curve has to be integrated over the entire
range ,of the initial size distribution curve. This can be accom-
plished either mathematically on a programmable calculator or
graphically by plotting penetration versus percent mass undersize
over the initial size range. Then the area under the curve
represents the total penetration, PT, of the system. The total
efficiency of the system can then be determined as:
E = 1 - PT (7-12)
92
-------
where Pt = overall penetration of the system, fraction
E = overall efficiency of the system, fraction
This value represents the predicted efficiency of the F/C
scrubbing system.
Limitations of The Design Model
The revised model is much simpler to use than our previous
version and it appears to give very good predictions. It also
offers the opportunity for easy modification to suit specific
situations. The limitations of the design model are given below:
1. The quencher unit is considered as a humidification
chamber but particle collection in this unit of the F/C
scrubbing system is neglected. The quencher would have an
effect on the collection efficiency of the larger particles
but the collection efficiency of the submicron particles would
be very low. Since the main area of difficulty is the removal
of submicron particles, the effect of particle collection across
the quencher is not amended to the design model.
2. The collection mechanisms of inertial impaction and
diffusiophoresis are the only ones considered. The flux force
theoretical equations for collection by thermophoresis and
centrifugation were previously presented in earlier studies
(Calvert, et al. 1973 and 1975) but the effect of inertial impac-
tion and diffusiophoresis would overshadow the effect of these
other mechanisms.
3. A constant value is used for foam density of the
froth over the sieve plates in the scrubber. In actuality,
there would be some variation due to different plate
hydrodynamics.
4. Particles are assumed to be wettable, but insoluble.
If the emitted particles are soluble in water, as was the case
at the demonstration plant, the expected performance for the
system would be better. The solubility of the particles in
water would depress the vapor pressure at the particle-gas
interface, resulting in nucleation at a lower saturation
ratio and more growth at a given saturation ratio.
93
-------
5. A constant, average value was used for the fraction
of water vapor condensing on the particles. This approxima-
tion of "f " is based on computer modeling runs using the
unit mechanisms of heat and mass balance at conditions comparable
to F/C scrubbing at the demonstration plant. This value is
affected by different process conditions, among which the impor-
tant ones are the liquid bulk temperature and the actual value
of liquid phase heat transfer coefficient.
6. The liquid phase heat transfer coefficient, hr, was
evaluated from literature correlations for heat and mass
transfer to spheres and liquid phase controlled mass transfer
on sieve plates.
7. The design model also assumes that all condensation
and particle growth occurs at the first plate of the scrubber
unit. The following plates then see the fully grown particles
which are larger and easier to collect. This is close to what
has been observed experimentally.
Example Calculation
An example calculation of particle penetration as pre-
dicted from the new design model is given in Appendix "F".
94
-------
CHAPTER 8
ENGINEERING ANALYSIS
As a framework for the engineering analysis of an F/C
scrubbing system applied to a secondary metal recovery furnace,
a near optimum design has been prepared and is discussed in
this chapter. Design refinements of the demonstration plant
are both possible and necessary because of several factors
including:
1. The demonstration plant was over-designed in order to
provide the apparatus and operational variations and the
extensive instrumentation necessary for a quantitative investi-
gation. A commercially practical prototype can be smaller,
simpler, and less expensive so a realistic view of F/C
scrubbing should be based on this latter approach.
2. Knowledge of the complexities and characteristics of
the source operation increased during the demonstration program.
Because of the great variability of the source and the conse-
quent demands on the control system, the design must be made
appropriate for the range of parameters as now known.
3. In the course of analyzing the demonstration plant
performance data and interpreting the complexities of operational
behavior, our basic understanding of F/C scrubbing in practice
has grown. Part of the increased engineering capability was
the development of a greatly simplified method for performance
prediction. An F/C system design can be optimized much more
readily now than was previously possible.
The following chapter presents a refined F/C design for a
secondary metal recovery furnace control system. Capital and
operating costs have been estimated for the F/C system and are
discussed in relation to a conventional scrubbing system.
95
-------
_Characteristics
The first matter to he resolved before proceeding with the
engineering design is the definition of the source operation
which is to be controlled. While the incinerator furnace
design and operation are important in determining the quantity
and nature of the emissions, their optimization is beyond the
scope of this study. For the present design we will assume
that the furnace and afterburner system will be such as to
produce emissions comparable to those encountered in the
demonstration program.
The type of wire or scrap being fed is the main factor
determining the severity of the emissions problem. Economics
and scrap supply enter into the picture and dictate what kind
of scrap is available and profitable to treat. Although one
cannot predict the market, the operator's preferences can be
considered. It was not possible for the source operator to
"burn" any grade of copper wire scrap, not even no. 1 premium
wire, and comply with the air pollution codes.
It would be adequate if the operator could burn all types
of no. 1 wire scrap although he would like to be able to burn
no. 2 and no. 3. It would not be sufficient to burn only no. 1
premium under present market conditions. Most of the operation
during the demonstration plant program was with no. 1 regular
and no. 1 premium wire so most of the data are on these.
We can set forth the criteria for burning no. 1 regular
wire (which will exceed the requirements for no. 1 premium)
and design an F/C system to satisfy them. The limited data
available on no. 2 and no. 3 wire indicate that a system
adequate for no. 1 regular wire would probably suffice for them
too. This is because the emissions from no. 2 and 3 wire are
higher than from no. 1 regular wire mainly because of increased
acid gases (impinger catch). The acid gases can be absorbed
with very high efficiency and it. is relatively simple and
inexpensive to provide this in the system for no. 1 wire.
The plot plan of the demonstration plant was dictated by
the operator's need for access routes and by the arrangement
-------
of equipment and scrap piles. While this is a special case,
it is likely that any commercial operation will have similar
problems. Consequently, the F/C system should be designed to
fit the plant encountered in the demonstration.
Liquid waste disposal requirements were obtained from the
local authorities and the demonstration plant was operated in
compliance. The major requirements are that the liquid dis-
charge to the sewer be neutral and not interfere with the sewage
treatment plant process. The small quantity and the composition
of the liquid waste from this F/C system are such that compliance
can be accomplished by neutralization and daily discharge of the
scrubber liquor.
Performance Criteria
The levels of particulate emissions are first established
in order to complete the overall efficiency necessary to
attain the desired outlet loading for compliance with the
legal pollutant limit from the furnace.
The inlet mass particuate loading measured during the
operation of the demonstration unit was 1.6 g/DNm3 plus an
approximate average of impinger loading of 0.15 g/DNm3. The
total value of 1.75 g/DNm3 was computed from the loading data
presented in Table 6-3 for modes A to F, which exclude peak-
to-peak data and uremium wire data. The value of 0.15 g/DNm3
was approximated from all the impinger data (presented in
Table "D" of the appendix) taken during the operation of the
demonstration unit.
The design outlet mass particulate loading was set at
0.08 g/DNm3. This was based on the legal limit of 0.10 g/DNm3
(^corresponding under average conditions to 0.1 gr/SCF corrected
to 12% C02) minus the expected entrainment from the scrubber
and cooling tower units as designed. From the inlet and outlet
design criteria, the overall fractional penetration required
was calculated to be 0.05. Thus, the system is designed to
give an overall efficiency of 95%.
The particulate characteristics were based on both the
demonstration unit data and the 1.4 m3/min pilot plant data.
97
-------
The size was approximated conservatively for the design of the
optimum F/C unit as follows:
d =0.75 ymA
r &
°g - 2-5
In addition, the system must be designed to neutralize and
efficiently remove the acid gas by mass transfer in order to
both reduce the emission level and prevent excessive corro-
sion of the system. The design criteria are summarized in
Table 8-1.
Process Description
The general characteristics of a refined F/C scrubbing
system suited to the design criteria are shown in Figure 8-1,
a schematic flow diagram. Major equipment items are also
identified in Figure 8-1, while stream quantities and condi-
tions are given in Table 8-2. The discussion which follows
will describe the design logic and the most important features
of the equipment and the operating conditions.
Flue gas, stream 1, will be drawn from a branch on the
afterburner stack into a saturator which serves to quench and
saturate the gas before it enters the cross-over duct. The
afterburner stack is fitted with a lid which can be opened
when the scrubber system is not operating or in case the
saturator liquid supply fails. The saturated gas flows through
the cross-over duct into a perforated plate condenser unit.
There moisture condensation will cause particle growth, removal
of particulate matter by diffusiophoresis, and removal of acidic
gas by absorption. Next the gas will pass through a venturi
scrubber and be vented to the atmosphere through a stack.
Water from the saturator drains to a holding tank and is
recirculated to the saturator unit. Liquor from the condenser
will be pumped to a cooling tower for evaporative cooling and
recycling. The liquor from the venturi scrubber will be
recycled through the cooling tower tank.
The liquor used to quench and scrub the gas will be
treated periodically with sodium carbonate to adjust the pH
98
-------
TABLE 8-1. DESIGN CRITERIA SUMMARY
Source
Charging rate (assumed)
Type of scrap
Plant layout
Operation schedule
Gas Flow
Gas emission rate
Gas temperature
Gas humidity
Contaminant Emissions to F/C
Acid gases
Particle concentration
(without impinger catch)
Particle mass median diameter
Particle geometric standard deviation
Particle impinger catch
Emissions from F/C
Scrubber (without entrainment)
Entrainment (scrubber and cooler)
Total particulates
Liquid discharge
1,800 kg/hr
No.l regular wire
As in demonstration
300 day/yr, 8 hr/day
200 Am3/min
700°C
0.06 g H20/g D.G.
0.85 kg mole
equivalents/day
1.6 g/DNm3
0.75 ymA
2.5
0.15 g/DNm3
0.08 g/DNm3
0.02 g/DNm3
0.1 g/DNm3
once/day, neutralized
99
-------
o
CD
Spray
Satura-
tor
-cH-
Venturi
Scrubber
4
1 1 1 i 1 I 1
Entr.
Sep.
5
.111 it
12
iinim
I
Mill I I I I I
1
Plate
Condenser
Soda Ash
Saturator
Settling
Tank
J
Condenser
Tank
14
H
Exhaust
Gas
mini
Liquid
Cooling Tower
Exhaust
i
Cool ing
Tower
Cool ing
Tank
I
15
Tower
•Drain
Ambient
Air
r
Make-
up
Water
Figure 8-1. Flowsheet for F/C scrubber optimum design at metals recovery furnace.
(See Table 8-2 for description of process streams)
-------
TABLE 8-2.
PROCESS STREAMS FOR F/C SCRUBBER OPTIMUM DESIGN
AT METALS RECOVERY FURNACE. (See Figure 8-1)
Stream
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Composition
Flue Gas
Sat. Gas
Cool Gas
Scrubbed Gas
Clean Gas
Exhaust
Hot Water
Hot Water
Make up Water
Cold Water
Hot Water
Cold Water
Cold Water
Hot Water
Cold Water
Temperature •
°C
700
73
36
36
36
36
49
73
21
32
49
32
32
49
32
Volume
Flow
*m3/min
H /min
200.0
100.0
61.5
69.1
69.1
61.5
56.7
38.1
736.0
755.9
123.0
"123.0
737.3
714.9
Mass
Flow
kg/min
70.3
88.7
68.8
68.8
68.8
68.8
56.7
38.1
736.0
755.9
123.0
123.0
737.3
714.9
Enthalpy
kgcal/kg
225
225
32.7
32.7
32.7
32. 7
49.0
73.0
21.0
32.2
49.0
32.2
32.2
49.0
32.2
Humidity
gH20/gD.A.
0.06
0.34
0.04
0.04
0.04
0.04
Gage
Pressure
cm W . C .
0
-5
-35
-114
-115
0
Gas flow rates in m3/min, liquor flow rates in £/min
-------
of the liquor to 7.0. The saturator tank with a total retention
time of 30 minutes also serves a a clarifier by settling out
some of the suspended solids. A batch discharge of 8,316 liters
at pH = 7.0 from all the tanks to the sewer is provided daily,
limiting the solids concentration to 1% maximum.
As indicated in Figure 8-1, the optimum design consists
of four major vessels which included the saturator, condenser,
venturi scrubber, and the cooling tower. The general design
considerations and brief mechanical description of each major
vessel are given below.
Saturator
The saturator cools and saturates 200 Am3/min (3,533 ACFM)
initially at 700°C to 73°C (163°F) by evaporating hot water
adiabatically. Acid gas absorption and neutralization also
occur to some degree in the saturator.
The saturator (see Figure 8-2) consists of a horizontal
cyclindrical spray chamber (0.91 m diameter x 1.52 m length).
The shell material selected is Hastelloy for protection against
high temperature and corrosion. The unit is located adjacent
to the stack and has a safety pressure switch to cause city
water injection into the unit in case of saturator pump
failure. This and the interlocked stack cap opening mechanism
will prevent hot, unquenched gas from entering other sections
of the F/C system.
The saturator spray requirement is 56.7 £/min (* 3 times
evaporation rate) of recirculated liquor which is pumped at
2.8 kg/cm2 (40 psig) to 10 spray nozzles. A holding tank
measuring 1.8m x 1.2m x 0.61m is specified for the saturator
and a self-priming centrifugal pump is used for pumping the
liquor to the unit. The external piping is chlorinated poly-
vinyl chloride (CPVC) while the internal piping is stainless
steel.
From the saturator the gas flows a distance of about
9.2 m (30 ft) through 0.46 m diameter fiberglass reinforced
plastic (FRP) duct to the condenser. The long distance
between the two units was necessary at the demonstration plant
102
-------
Gas Inlet
200 m
§ 700°C
Furnace
Stack
Hastelloy Shell
(91 cm dia. x 152 dm length)
Spray
Manifold
Outlet Duct
(46 cm dia.)
I
Gas Outlet
100 mVroin
@ 73°C
t
Water Inlet
56.7 £/min
@ 49°C
Water Drain
to Saturator Tank
ELEVATION VIEW OF SATURATOR
Figure 8-2. Saturator for F/C optimum design.
103
-------
site in order to give clear access to the incinerator furnace
and is provided for in the refined design. While there is
some risk of burning the FRP duct in case of a failure of the
pump and the safety interlocks, the cost is much less than
corrsion resistant metal ducting and a heavier support
structure.
Condenser
One purpose of the condenser unit is to cool the saturated
gas at 73°C down to 36°C and a flow rate of 61.5 Am3/min (2,171
ACFM). Another major function of the condenser is to provide
sufficient mass transfer capacity to absorb substantilly all
of the acid gases which do not condense. This cooling and
absorption is accomplished by bringing the gas in countercur-
rent contact with cold water at 32°C. The condenser is designed
for a condensation ratio of 0.30 g H20 dry which causes
sufficient growth to occur. The amount of particle growth
was predicted by means of the design methods described in
Chapter 7 to satisfy the design criteria for this refined
system. Figure 8-3 is a log probability plot of the original
particle size distribution and the size distribution after
condensation and growth. The use of these data for the
prediction of scrubber system performance is discussed later.
The condenser (see Figure 8-4) is a vertical counter-
current tower with three l.lm diameter plates and 2.4m
height. The shell is constructed of FRP and the sieve trays
are made out of stainless steel.
The water requirement of the condenser is 736 £/min at
32°C. The water is introduced at the top and cascades down
to the bottom plates, maintaining cross flow on the sieve
plantes. A self-priming centrifugal-type pump at 750 £/min
and 2.5 kg/cm2 head and a polypropylene holding tank, mea-
suring 1.8m x 1.8m x 0.9m are specified.
Venturi Scrubber
Particle growth, particle deposition by diffusiophoresis
and thermophoresis, absorption of acid gases, and removal
of particles larger than a few microns aerodynamic diameter
104
-------
3.0
Initial Conditions
= 0.75 ymA, a = 2.5
n = 1. 2 x 10 9 /DNcm
0,1
10 20 30 40 50 60 70 80 90 95 98 99
MASS % UNDERSIZE
Figure 8-3. Predicted grown particle size distribution for F/C
scrubber optimum design.
105
-------
Gas Inlet
99 niVmin
@ 73°C
Inlet Duct
(46 cm dia.)
later In"
'36 A/min
32°C
;t
din
r
^
}
./
i 36°C
__ |
1
1
i
_| Jl.
1
j,
l •
.J 1
A73°C
T
^pj
t
61
*
1
46
-t
46
91
1
I
Outlet Duct
( 30.5 cm dia.)
Gas Outlet
6.15 mVmin
e 36°C
91 cm ELEVATION VIEW OF
CONDENSER
3 sieve plates (316 SS)
107 cm diameter shell
(FRP)
Water Out
Figure 8-4. Condenser for F/C optimum design.
106
-------
will occur in the condenser. The purpose of the venturi
scrubber is to provide the capability for collecting sub-
micron particles, which it can do more efficiently than a
sieve plato scrubber. On the other hand, the .scive plate
scrubber is much more efficient than the venturi for mass and
heat transfer. Consequently, while it would be appealing to
use either one or the other type of scrubber from the stand-
point of simplicity the design criteria -tre such as to make
it best to use both.
The venturi scrubber will be followed by a tube-bank-type
entrainment separator. This type of separator was selected on
the basis of the small cut diameter required to control the
predicted entrainment rate and size distribution. Some
additional mass transfer capability is provided by the combi-
nation of the scrubber and entrainment separator.
The venturi section is designed with a pressure drop of
79 cm W.C. and a Qj/Qp ratio of 2 £/m3. The required penetra-
tion across the system was established at 0.05. Accounting
for collection by diffusiophoresis in the scrubber, the
required penetration through the venturi was calculated as;
0.74
The grown particle size which enters the venturi unit is given
in Figure 8-4. A penetration-particle diameter relationship
was calculated for different pressure drops and integrated
over the entire size distribution. By trial and error, a
pressure drop of 79 cm W.C. was calculated as necessary to
obtain a penetration of 0.068 when integrated over the size
distribution range. The penetration grade curve for the ven-
turi at this pressure and an overall penetration of 0.068 is
given in Figure 8-5.
Physically, the venturi is represented in Figure 8-6.
The venturi is built from 1.3 cm wall thickness FRP with a
10 7
-------
o
I— I
H
OS
Pn
0.3
0.2
0.1
0.01
INITIAL PARTICLE DIAMETER (d ), ymA
Pa
Figure 8-5. Predicted penetration for venturi
scrubber as designed for optimum
system.
108
-------
Gas Inlet
(61.5 mVmin @ 36°C)
Venturi | 1 I
Section \ 1
(FRP) 2' \ »
t \ I/
Tube Bank
Entrainment
Separator ,
S I ^^f
let \
Water
Drain
, -^4 2.9 |-^-
W 1 cm 1
'••- O O U (
y
T
D(
Inlet Duct
(30.5 cm dia.)
Water Inlet
(123 £/min @ 32°C)
Throat (14.4 cm dia.)
ELEVATION VIEW .OF
SCRUBBING SECTION
TOP VIEW OF TUBE ARRANGEMENT
OF ENTRAINMENT SEPARATOR
5 stages
O OOOOO
oooooo
o oo o o o
Figure 8-6. Venturi scrubber for F/C optimum design.
109
-------
gas inlet of 30.5 cm ID and a throat diameter of 12.7 cm ID.
The liquor requirements are 123 Jl/min, which corresponds to a
liquid-to-gas ratio i 2 £/m3. A self-priming centrifugal pump
giving an operating flow of 132 £/min at 2.5 kg/cm2 pressure
is provided for this purpose. The pump inlet is fed from the
cooling tower sump and PVC piping is specified for both suction
and pressure sides of the piping.
The second section of the scrubbing unit is a tube bank
entrainment separator with a cut diameter of 13 ym ard a
designed pressure drop across the unit of 0.8.cm W.C.
The inlet size distribution of the drops to the entrain-
ment separator was approximated using the Nukiyama and
Tanasawa correlation:
/o \ 1-5
, - 50
r \.^J"/ •>cv-} \ G
where d = Sauter mean diameter of drops, cm
Up = air velocity relative to drops, cm/sec
QT = water flow rate m3/sec
Qp = air flow rate m3/sec
Using equation 8-1 the mean Sauter diameter of the drops was
estimated to 163 ym. As discussed by Calvert et al. (1975),
the Sauter mean diameter is typically 70% to 90%. of the mass
median diameter. The conservative value of 70% of the mass
median diameter was used. Thus, the mass median diameter
was calculated to be 233 ym with a corresponding "a "
calculated to be 2.3.
The design of the entrainment separator was based on an
inlet entrainment of 2,200 cm3/DNm3, calculated from a L/G
ratio of 2 £/m3 and specific gravity of 1.1 g/cm3. The
minimum outlet from the entrainment- separator was set at
0.01 g/DNm3 with a corresponding entrainment o£ 0.90 cm3/DNm3
The cumulative entrainment was plotted versus drop diameter,
and the cut diameter was approximated to be 13 ym. Figure
8-7 gives the cumulative entrainment as a function of drop
diameter.
110
-------
4
Gas Inlet
(61.5 mVmin @ 36°C)
Venturi f 11
Section \ 1
(FRP) 2' \ 1
t \ l>
Tube Bank
Entrainment
Separator ,
)
Inlet Duct
(30.5 cm dia.)
Water Inlet
(123 £/min § 32°C)
Throat (14.4 cm dia.)
ELEVATION VIEW .OF
SCRUBBING SECTION
Water
Drain
r
TOP VIEW OF TUBE ARRANGEMENT
OF ENTRAINMENT SEPARATOR
"O O O O O O
o ooooo
o o o o o o
o ooooo
Vertical Tubes -
5 stages
Figure 8-6. Venturi scrubber for F/C optimum design.
109
-------
gas inlet of 30.5 cm ID and a throat diameter of 12.7 cm ID.
The liquor requirements are 123 £/min, which corresponds to a
liquid-to-gas ratio i 2 i/m3. A self-priming centrifugal pump
giving an operating flow of 132 £/min at 2.5 kg/cm2 pressure
is provided for this purpose. The pump inlet is fed from the
cooling tower sump and PVC piping is specified for both suction
and pressure sides of the piping.
The second section of the scrubbing unit is a tube bank
entrainment separator with a cut diameter of 13 ym ard a
designed pressure drop across the unit of 0.8 cm W.C.
The inlet size distribution of the drops to the entrain-
ment separator was approximated using the Nukiyama and
Tanasawa correlation:
where d = Sauter mean diameter of drops, cm
Up = air velocity relative to drops, cm/sec
(X = water flow rate m3/sec
Qp = air flow rate m3/sec
Using equation 8-1 the mean Sauter diameter of the drops was
estimated to 163 ym. As discussed by Calvert et al. (1975),
the Sauter mean diameter is typically 70% to 90i of the mass
median diameter. The conservative value of 70% of the mass
median diameter was used. Thus, the mass median diameter
was calculated to be 233 ,ym with a corresponding "a "
calculated to be 2.3.
The design of the entrainment separator was based on an
inlet entrainment of 2,200 cm3/DNm3, calculated from a L/G
ratio of 2 £/m3 and specific gravity of 1.1 g/cm3. The
minimum outlet from the entrainment separator was set at
0.01 g/DNm3 with a corresponding entrainment ,of 0.90 cm3/DNm3.
The cumulative entrainment was plotted versus drop diameter,
and the cut diameter was approximated to be 13 ym. Figure
8-7 gives the cumulative entrainment as a function of drop
diameter.
110
-------
10.0
0.5
50
15 20 25 30
DROP DIAMETER, ym
Figure 8-7. Assumed cumulative scrubber entrain-
ment versus drop diameter for
optimum system.
Ill
-------
Physically, the entrainment separator (see Figure 8-6)
consists of 6 rows of 1.9 cm diameter PVC tubing with a center-
to-center spacing of 2.9 cm. The expected velocity through the
rectangular orifice is about 13 m/sec with a corresponding
pressure drop of 0.8 cm W.C.
The entrainment separator is oriented with the tubes
running in the vertical direction and enclosed in a 35cm x 35cm
x 35cm housing. The unit is provided with a drain which flows
to the cooling tower sump for recirculation to the scrubber.
Cooling Tower
The purpose of the cooling tower is to reduce the
temperature of the liquor from the condenser at a flow rate
of 736 £/min from about 49°C to 32°C. This is accomplished in
a mechanical induced draft cooling tower by means of evaporative
cooling. An induced draft of ambient air is moved through
the unit by a fan located at the top. The liquor flow is intro-
duced at the top and flows down the fill area (see Figure 8-8).
A spray-type cooling tower was used in the demonstration
plant because of concern for the fouling of a filled cooling
tower if it were used. Experience showed that solids deposi-
tion in tanks and wall surfaces of the cooling tower was no
problem. Any deposits could be easily washed off. The spray-
type tower does have the disadvantages of requiring a higher
pressure pump and a more efficient "drift" entrainment sepa-
rator than a filled tower.
The total cooling requirements of the cooling tower is
approximately 880 kJ/s (50,000 BTU/min). The liquor flow
rate is 736 £/min, for which a self-priming centrifugal
pump with 2.5 kg/cm2 head is specified.
The cooling tower is so designed to limit the entrainment
to 0.006% of the circulating liquor rate. This is approximately
0.010 gram of emission per DNm3 of flue gas through the
scrubber, with the result that the total emission from the
system is kept below 0.1 g/DNm3.
11.
-------
I
Exhaust
Air
Hot Water In
736 £/min 49°C
7 1/2 HP
Fan
ENTRAINMENT SEPARATOR
I I I II I M I I t T I I I I I II
INDUCED DRAFT COOLING TOWER
880 kJ/s (50,000 BTU/min)
Cold Water
Ambient
<*- Air
Water Outlet
715 £/min, 32°C
Figure 8-8. Cooling tower of F/C optimum des
recovery furnace.
ign at metals
113
-------
COST OF OPTIMUM SYSTEM
This section discusses the economic aspects of F/C
scrubbing by estimating the capital and operating cost of an
F/C scrubbing unit at the metals recovery furnace. The economic
data are based on the optimum design as presented in the
previous section and assumes an overall efficiency of 95%
over quite unfavorable particulate conditions (d = 0.75 ymA
IT &
and a = 2.5) . '
Capital Cost
The capital cost of an F/C scrubber unit would consist of
total equipment cost and other direct cost such as installation,
piping, electrical, etc. along with indirect costs such as
engineering, construction overhead, contingencies, etc. The
total equipment costs are based on quoted prices while the other
direct and indirect costs have been estimated using ratio factors
based on delivered-equipment cost. (Peters et al., 1968)
The detailed equipment costs are presented in Table 8-3
and the other direct and indirect costs are given in Table 8-4.
In summary the following costs are estimated for the optimum F/C
scrubber at the metals recovery furnace:
Direct Costs
Equipment $ 25,290
Piping, Instruments, etc. 51 ..084
$ 76,374
Indirect Costs
Engineering, contingencies, 34,647
etc.
34,647
Total capital investment $111 021
Operating Costs
The operating cost for the optimum design consists of the
annual cost of utilities (power and water), raw materials, and
maintenance. The computed costs are summarized in Table 8-5.
The utilities cost is based on both power and water
usage. Power usage was estimated to be 42 kw-hr, 8 hours
114
-------
TABLE 8-3. TOTAL EQUIPMENT COST ESTIMATE
FOR F/C OPTIMUM DESIGN
EQUIPMENT
UNIT COST, $*
TOTAL COST, $
Venturi Scrubber
Venturi Section
Entrainment Separator
Pump, Motor
Total
Cooling Tower
Cooling Tower, Fan,
Motor, Sump
Pump, Motor
Total
Condenser
Shell (fiberglass)
Trays
Tank
Pump, Motor
Total
Saturator
Shell (Hastelloy)
Tank
Pump, Motor
Safety Switches, etc.
Total
Blower § Motor
Blower
Motor
Total
TOTAL EQUIPMENT COST
325
600
770
7,300
1,590
1,000
950
500
1,590
2,500
500
1,065
1,000
4,515
1,085
1,695
8,890
4,040
5,065
5,600
25,290
*Cost based on 4th quarter, 1976,
115
-------
TABLE 8-4. DIRECT AND INDIRECT COST ESTIMATE
FOR F/C OPTIMUM DESIGN
TYPE OF COST
Direct
Equipment
Installation
Instruments
Piping, Ducting
Electrical
Site Preparation
Structural
TOTAL DIRECT COST
Indirect
Engineering
Construction Overhead
Contractor Fee
Contingency
TOTAL INDIRECT COST
TOTAL CAPITAL INVESTMENT
RATIO
1.00
0.47
0.18
0.66
0.11
0.10
0.50
3.02
0.33
0.41
0.21
0.42
1.37
4.39
COST, $
25,290
11,886
4,552
16,690
2,782
2,529
12,645
76,374
8,346
10,369
5,310
10,622
34,647
111,021
»
116
-------
TABLE 8-5. OPERATING COST OF OPTIMUM F/C DESIGN
TYPE OF COST
Maintenance
Water Usage
Power
Raw Materials
UNIT COST
0.06 of total capital
investment
$0.034/1,000 liters
$0.045/kw-hr
$12.25/100 kg
COST
6,660
180
4,520
1,650
TOTAL ANNUAL OPERATING COST $13,010
117
-------
operation, 300 operating days per year. The water usage is
approximately 17,780 £/day, mainly due to evaporation and
blowdown losses, and unit cost of water was based on $0.034/1,000
liters.
Raw material cost is mainly the use of sodium carbonate as
the neutralizing agent. Based on the demonstration plant opera-
tion, approximately 45 kg/day is necessary at a cost of $12.25/
100 kg. The maintenance cost was estimated as a ratio of 0.06
of the fixed capital investment.
The total annual cost of the F/C scrubber at the metal
recovery furnace was calculated as the sum of the annual
capital cost and the annual operating cost. The annual
capital cost is based on a 10-year life and straight-line
depreciation and calculated as $11,102. The annual
operating cost as shown in Table 8-4 is estimated at $13,010.
Thus, the total estimated annual cost of a F/C scrubber adds
up to $24,112.
CONVENTIONAL SCRUBBER
For comparison with the F/C scrubbing system a conventional
high energy system is considered below. It will be seen that
conventional scrubbing would be impractical for no. 1 regular
wire because of the excessively high pressure drop required.
Thus, a comparison could not be made within the bounds of the
design criteria but in order to illustrate the potential advan-
tage of an F/C system the data are given for no. 1 regular and
no. 1 premium wire burning control.
Any pollution control system on this source would have to
reduce the acid gas emission rate as well as control the condens-
able particulates. Consequently, the high energy scrubber would
have to include the same mass transfer capacity as the F/C sys-
tem. The arguments in favor of using a saturator right at the
stack in order to reduce the actual gas volumetric flow rate
and temperature and to enable the use of FRP also apply to the
conventional scrubber case.
118
-------
The conventional scrubber system would have the same
components as the F/C system except for the cooling tower and
associated apparatus. The scrubber and sieve plate column
tanks can be combined into one large tank. Thus, the flow
sheet for the conventional scrubber will be similar to that
for the F/C system.
In order to achieve the penetration of 0.05 on an inlet
d = 0.75 ymA and a = 2.5, a pressure drop of 535 cm W.C.
(210 in. W.C.) would be required. This would be an imprac-
tically high pressure drop for a fan and would involve
operating at a suction pressure of 368 mm Hg absolute if an
induced draft compressor were used.
Beside the high pressure drop, the fan for a conventional
scrubber would have to treat a larger gas volume than the F/C
scrubber fan. For example in this case, even if the fan had
the same pressure drop the F/C fan would only require 61.5%
as much power as the conventional. For comparison, the F/C
system handles 69 Am3/min at 36°C and a pressure drop of -138
cm W.C. The gas leaving a conventional scrubber would be at
the adiabatic saturation condition corresponding to the inlet
gas enthalpy so the fan would have to handle 112 Am3/min at
73°C for a similar pressure drop of -138 cm W.C.
Equipment costs for a conventional high energy scrubber
to control no. 1 regular wire emissions are estimated to be
as shown in Table 8-6.
The estimated total annual cost for the F/C system to
control no. 1 regular wire burning was given previously as
$24,112. It is obvious that even if a conventional high energy
scrubber could be made to operate at AP = 535 cm W.C., it
would not be worth doing.
Conventional Scrubber for No. 1 Premium Wire
A conventional scrubber could be operated to control the
emissions from burning no. 1 premium wire at a high but prac-
tical pressure drop. The penetration required for no. 1 pre-
mium wire recovery is 0.12, based on an inlet particle load-
ing of 0.68 g/DNm3 and an outlet requirement of 0.08 g/DNm3.
119
-------
TABLE 8-6
HIGH ENERGY SCRUBBER COSTS FOR REGULAR NO. 1 WIRE
Venturi Scrubber fAP = 535 cm W.C., QL/QG = 2 £/m3)
Venturi Section (FRP) $ 325
Entrainment separator (PVC and FRP) 600
Pump and motor 860
Sub-total $ 1,785
Condenser (same as F/C) 4,038
Sieve plate column (same as F/C) 5,065
Blower (ficticious) 30,703
Motor (ficticious) 10,840
Total Equipment 52,431
Total Capital Investment $230,172
Total annual costs of operating the 535 cm W.C. scrubber
system would be as follows:
Capital cost, 10 years straight line
depreciation $23,020
Maintenance 13,810
Water 180
Raw Material 1,650
Power 39,400
Total Annual Cost $78,060
120
-------
The particle size distribution was taken as the same as for
no. 1 regular wire recovery, i.e., d = 0.75 ymA and a = 2.5,
± £:> G)
An F/C scrubber system would need an overall pressure drop of
70 cm W.C. while a conventional scrubber would require 238
cm W.C. to give 12% penetration.
The main costs for the two systems are given in Table 8-7,
It can be seen that F/C scrubbing to control no. 1 premium
wire recovery would have an annual cost about 66% of that for
conventional high energy scrubbing. As previously discussed,
however, the system would have to be capable of controlling
emissions from no. 1 regular wire in order to be commercially
useful.
121
-------
TABLE 8-7. COST COMPARISON FOR PREMIUM WIRE RECOVERY
Cost Cost for
Cost Item for F/C Conventional
Venturi $ 1,700 $ 1,790
Cooling Tower 8,900 0
Condenser 4,040 4,040
Saturator 5,060 5,060
Blower § Motor 3,790 14,450
Total Equipment 23,490 25,340
Total Capital $103,121 $111,242
Investment
Depreciation $10,310 $11,125
Maintenance 6,185 6,675
Water 180 180
Raw Materials 1,650 1,650
Power 2,370 11,530
Total Annual Cost $20,695 $31,160
122
-------
CHAPTER 9
FUTURE RESEARCH RECOMMENDATIONS
In the course of this demonstration plant program
there has been a considerable gain in our understanding of
F/C scrubbing in addition to the general validation of
the system in a practical setting. Some of the lessons
learned have changed previous notions and point up the need
for new information, while others reinforced our former percep-
tions of research needs.
Our concept of F/C scrubbing has been sharpened since
this demonstration began. We now see the bare essentials of
F/C scrubbing as follow:
1. The fine particle collection efficiency (fractional)
due to diffusiophoresis will be about equal to 85% of the
volume fraction of gas condensed, regardless of particle size
and equipment type.
2. Thermophoretic deposition will add only a few percent
to the particle collection efficiency by diffusiophoresis.
3. High efficiency fine particle collection must be
accomplished by some mechanism other than flux force deposi-
tion. If the high efficiency is to be achieved at lower
pressure drop (or energy input) than a conventional (no F/C)
scrubber, there must be particle growth by water condensation.
The condensation must occur before the gas enters the high
efficiency section of the scrubber system.
Some of the future research needs discussed below are
for the purpose of providing information related to the points
above. Others of the needs are concerned with commercial
operating systems and their accessory components. Progressing
from the basic through the applied, our recommendations for
123
-------
future research are as discussed below.
Particle Growth
A
Laboratory research on particle growth is necessary to
provide data which can be used in validating and/or revising
the mathematical model. The experiments should be done with
wettable and non-wettable particles in several types of
bench-scale apparatus which incorporate contacting mechanisms
typifying large scale equipment.
This experimental and analytical work is needed to clear
up the present uncertainties about several interacting
phenomena as represented in the mathematical model for
particle growth. The points needing clarification are as
follow:
1. The nucleation of condensation on the surface of
insoluble particles may require some supersaturation of the
gas, depending on the wettability of the surface. As
presently set, the model accounts for condensation and
growth when the saturation ratio is 1.0 or greater. It has
been assumed that the supersaturation which occurs in the
gas phase boundary layer close to the cold liquid surface
when the bulk of the gas is just saturated (i.e., s = 1.0)
might be sufficient to nucleate condensation on slightly
wettable particles.
Because the degree and extent of the boundary layer
supersaturation effect depends on the conditions of the
gas and liquid, the geometry, and the hydrodynamics, there
is no simple relationship defining the "effective" saturation
ratio. As will be seen, it is also impossible to distinguish
between the effects of the several phenomena which occur
simultaneously during condensation scrubbing.
2. Liquid phase heat transfer coefficient has a
pronounced effect on liquid surface temperature and, thereby,
on heat and mass transfer and gas phase saturation ratio.
There are large differences among heat transfer coefficients
computed from various correlations given in the literature
124
-------
for sieve plates and even larger differences between them
and some experimental data from earlier studies of F/C
scrubbing.
3. Gas phase heat and mass transfer coefficients are
somewhat better defined than for the liquid phase in sieve
plates but there is still appreciable uncertainty as to the
gas flow patterns and the size and shape of the gas-liquid
interface. These factors all have a bearing on the state
of saturation and mixing of the gas and particles.
4. The rate of condensation from the gas depends on the
temperature and vapor pressure differences between the phases
and on the transfer coefficients. It also affects the frac-
tion of total condensation which goes to the particles (i.e.,
"f ") as predicted by the mathematical model.
5. The particle number concentration also influences "f ",
as computed from the model. The experimental data taken
previously have not enabled very precise computation of
number concentration, so the influence of this parameter has not
been distinguishable from those of other parameters.
6. Soluble particles can cause condensation at satura-
tion ratios less than 1.0 because the vapor pressure of water
is lowered by the solute. The present demonstration plant
source particles contained a large fraction of soluble
material but the beneficial effect of this was neglected in
predictions computed with the model.
Field Measurements
Field measurements of particle number concentration and
growth characteristics for a> number of important air pollution
sources would provide the key information needed to assess the
suitability of F/C scrubbing for the specific source. Apparatus
of the same general size and nature as that used for making
performance tests could be taken anywhere in the U.S. and
operated by a 3-man crew.
By sampling the source plant effluent and measuring the
particle size after condensation and growth, one has evaluated
the combined effects of particle number concentration,
125
-------
solubility, wettability, and size distribution. Gas tempera-
ture and humidity effects will also be properly accounted for.
Liquid temperature and cooling apparatus design will be
the controllable parameters.
Once the measurements have been made and the aerodynamic
diameter of the grown particles determined, it will be a
routine matter to design a scrubber to give the appropriate
particle collection efficiency. As mentioned early in this
chapter, the diffusiophoretic deposition can be accounted
for with a simple computation. Thus, the main outlines of
an F/C scrubber system for a specific source could be estab-
lished with a good degree of confidence on the basis of
particle growth data.
The scale of the experiment should be based on a gas
flow rate on the order of 0.1 m3/min (3 CFM). This will give
a balance between being large enough to minimize wall losses
and small enough to require convenient sized accessories. The
sampling system would be very similar in nature to the systems
used for particle sampling and sizing in the plant.
A two or three-man crew^ could go to a plant site, set up
the equipment, and make the measurements in about one week,
exclusive of travel time. Special test conditions, measure-
ments in a number of system variations, or the aquisition of
other kinds of data could increase the time requirement.
By making the particle growth measurements in a number
of plants representing the major sources which are apparently
amenable to F/C scrubbing, one would obtain the essential data
for making reliable process designs and cost estimates for
controlling these sources by F/C scrubbing. Without this
experimental measurement one can only speculate on the basis
of several assumptions about the particles from a specific
source and the resulting design and cost estimate will be
very approximate.
Table 9-1, from Calvert et al. (1975), lists the major
fine particle sources for the U.S.A. with the features that
126
-------
NOTE:
IV.
to
--J
Table 9-1. MAJOR INDUSTRIAL PARTICULATE SOURCES
FOR WHICH F/C SCRUBBING IS ATTRACTIVE
The following information was taken from the Midwest Research Institute
Report (1970), The source number (Roman numeral) refers to its rank in
the U.S. as an industrial particulate pollution source.
NET
AMMTTAT CONTROL EMISSIONS
SOURCE FRACTION MKg/yr
VI.
IRON AND STEEL
A. Sinter Plants (Sintering
process)
B. Coke Manufacture
1. By-Product
2. Pushing § Quenching
C. Blast Furnace
D. Steel Furnaces
1. Open Hearth
2. Basic Oxygen
3. Electric Arc
E. Scarfing
FOREST PRODUCTS
A. Wigwam Burners
B. Pulp Mills
1. Kraft Process
a. Recovery Furnace
b. Lime Kilns
c. Dissolving Tanks
2. Sulfite Process
(Recovery Furnace)
3. NSSC Process
a. Recovery Furnace
b. Fluid-Bed Reactor
4. Bark Boilers
46,300,000 MKg of Sinter
81,600,000 MKg of Coal
82,800,000 MKg of Coal
80,600,000 MKg of Iron
59,700,000 MKg of Steel
43,500,000 MKg of Steel
15,200,000 MKg of Steel
118,800,000 MKg of Steel
24,900,000 MKg of Waste
22,000,000 MKg of Pulp
2,300,000 MKg of Pulp
756,000 MKg of Pulp
3,200,000 MKg of Pulp
1,100,000 MKg of Pulp
470,000 MKg of Pulp
0.90
0
.99
.40
.99
.78
.68
.91
.94
.30
.91
.91
.70
46,300
81,600
19,000
52,600
306,000
9,000
16,300
57,200
120,000
149,000
29,900
38,100
9,000
900
38,100
74,400
-------
TABLE 9-1 (Continued)
SOURCE
VII. LIME
A. Rotary Kilns
B. Vertical Kilns
VIII. PRIMARY NONFERROUS METALS
A. Aluminum
1. Calcining of Hydroxide
2. Reduction Cells
a. H. S. Soderberg
b. V. S. Soderberg
c. Prebake
h- B. Copper
TO 1. Roasting
2. Reverb. Furnace
3. Converters
C. Zinc
1. Roasting
a. Fluid-Bed
b. Ropp, multi-hearth
2. Sintering
3. Distillation
D. Lead
1. Sintering
2. Blast Furnace
3. Dross Reverb. Furnace
XI. ASPHALT
A. Paving Material
1. Dryers
2. Secondary Sources
ANNUAL
PRODUCTION
14,700,000 MKg of Lime
1,600,000 MKg of Lime
5,300,000 MKg of Aluminum
730,000 MKg of Aluminum
640,000 MKg of Aluminum
1,600,000 MKg of Aluminum
520,000 MKg of Copper
1,300,000 MKg of Copper
1,300,000 MKg of Copper
690,000 MKg of Zinc
138,000 MKg of Zinc
560,000 MKg of Zinc
560,000 MKg of Zinc
420,000 MKg of Lead
420,000 MKg of Lead
420,000 MKg of Lead
228,000,000 MKg of Material
NET
CONTROL
FRACTION
0.81
.39
90
40
,64
,64
,85
,81
,81
98
85
,95
.86
.83
.50
.96
.96
EMISSIONS
MKg/yr
267,000
3,600
52,600
31,700
9,000
18,100
6,000
25,400
29,900
13,600
3,600
2,700
13,600
15,400
9,000
1,800
150,000
36,300
-------
would be generally suited to F/C scrubbing. Given the particle
growth data, one could prepare another table showing the costs
of conventional and F/C scrubbing for attaining given levels
of emission rate. The engineering design and analysis
involved would be straightforward and based on existing
knowledge and methodology.
Cooling Towers
Field studies of solids deposition on surfaces of various
materials which could be used for cooling tower construction
would give the most important piece of information heeded in
selecting the type of cooler to use. These tests would be
done at the same sites as described in the previous section
and could most efficiently be done concurrently with the
particle characterization tests.
Prior concern over the possibility that solids deposition
on cooling tower surfaces would lead to the heavy buildup of
adherent scale led to studies of spray-type coolers. The
spray coolers have their own drawbacks which are sufficiently
serious that packed or filled cooling towers look attractive
again. Experience with the demonstration plant cooling tower
and observations of other systems in the field leads us to
believe that in some systems solids deposition on plastic
surfaces may cause no problems.
The cooling tower is an important part of F/C scrubbing
and represents the major cost difference between F/C and
conventional scrubbing for many applications. Therefore, the
use of a standard commercial cooling tower would give the
best combination of cost, reliability, and proven design
features. Purchase and installation of a standard cooling
tower are also routine matters which can be accomplished
through many vendors.
The solids deposition test involves the simple process
of pumping a scrubber liquor over pieces of various packing
materials to simulate their exposure in a cooling tower. If
scrubber liquor is not available some collected particulate
material can be mixed with water in the proper concentration
129
-------
and used in a recirculating system.
The length of the test period will depend on the severity
of the deposition problem, if any. Observation over a few
days of continuous operation should enable a rough evaluation
of the deposition characteristics. The experimental apparatus
can be left running for days or weeks with only occasional
attention. Thus, it could be set up at a plant and left in
the care of plant personnel for an extended test. Alterna-
tively, in some cases the particulate material can be shipped
to the contractors laboratory and the tests performed there.
It would be most economical and convenient to carry out
the solids deposition tests at the same time as particle
growth measurements at a given plant. A three-man crew could
set up the solids deposition experiment with no more than an
additional day's time if they were already at the plant site.
Steam Injection
The use of steam injection into saturated gas is an
attractive but insufficiently explored ramification of F/C
scrubbing. A bench-scale laboratory study could yield the
information needed to determine the optimum balance between
the quantity of steam to inject and the amount of condensation
by cooling. Engineering design studies followed by pilot
tests should be done to identify the best way of generating
steam inexpensively.
Data from previous studies indicated that fine particle
collection efficiency was greater when a given amount of
steam was introduced into the gas than when an equivalent
quantity was condensed from the gas. (See Calvert, et al.,
1975). The steam injection experiments were not made under
the same conditions so the comparisons between them and F/C
scrubbing with condensation only are not conclusive. However,
the apparent benefit of steam injection is so large that
further study is warranted.
The reason(s) for performance improvement by steam
injection is (are) not known but can be hypothesized. If
steam is mixed with saturated gas only a small fraction will
130
-------
condense, depending on gas temperature. Thus, a given quantity
of steam (say, 0.05 g/g dry gas) will give less condensate than
0.05 g/g, yet the particle collection efficiency is higher than
for just condensation of 0.05 g/g. The most persuasive
explanation is that steam injection causes an extremely high
saturation ratio in .the vicinity of the injection nozzle and
this enhances the nucleation of condensation. This mechanism
should be more significant for insoluble particles than for
soluble ones.
If it is possible to obtain a substantial benefit from
injection without subsequent cooling of the gas, there will be
a reduction of the liquor cooling cost. On the other side,
there will be some cost for steam generation. It would be
valuable to know what benefits could be obtained by using
various proportions of steam injection and condensation and
what the costs would be.
An experimental program of determining particle growth
under a range of parameters would provide the information
needed to predict scrubber performance. The experiments could
be done on bench scale with gas flow rates on the order of
0.1 to 0.5 m3/min. Soluble and insoluble particles should
be studied at number concentrations ranging from 106/cm3 to
109/cm3.
Costs for steam generation should be determined by
engineering analysis and design studies. Conventional and
non-conventional waste heat boilers should be evaluated.
Cooling costs can be based on existing technology.
Demonstration Plant
F/C scrubbing is highly suited to the control of fine
particles from basic oxygen furnace (B.O.F.) emissions and
a demonstration plant would be valuable. The B.O.F. emits
gas at temperatures up to around 2,000°C, which is unusually
high for a large volume source. The high gas enthalpy gives
the opportunity to evaporate a lot of water and to obtain a
high condensation ratio even with relatively hot scrubber
water.
131
-------
It is generally necessary to have a water-cooled hood
over the B.O.F. vessel mouth in order to protect the
ductwork and to reduce the gas temperature before quenching.
There is great concern that the gas temperature be less
than about 1,700°C when it is first contacted with water
so that dissociation of water will not cause explosive
mixtures of hydrogen and oxygen to form. Since steam can be
generated in the water-cooled hood, which must be provided
in any case, there is an opportunity to obtain steam at
very low cost.
B.O.F. emissions contain a large percentage of fine
particles, often running over 50 mass percent smaller than
1.0 ymA diameter. These emissions must be collected at
fairly high efficiency. Thus, the criteria for amenability
to F/C scrubbing are all satisfied by B.O.F. emissions.
Following the presently reported demonstration plant,
a second F/C demonstration program has been initiated on
a foundry cupola. The B.O.F. demonstration plant would make
a reasonable extension of F/C scrubbing technology into
extremes of temperature and magnitude.
132
-------
REFERENCES
1. Calvert, S., J. Goldshmid, D. Leith, and D. Mehta.
Scrubber Handbook. A.P.T., Inc., EPA Contract No.
CPA-70-95. NTIS No. PB 213-016. August, 1972.
2. Calvert, S., J. Goldshmid, D. Leith, and N. Jhaveri.
Feasibility of Flux Force/Condensation Scrubbing for
Fine Particulate Collection. A.P.T., Inc., EPA
Contract No. 68-02-0256. NTIS No. PB 227-307.
October, 1973.
3. Calvert, S., and N. Jhaveri. Flux Force/Condensation
Scrubbing. J. Air Pollution Control Association.
24 (10): 947-951. October, 1974.
4. Calvert, S., N. Jhaveri, and T. Huisking. Study of
Flux Force/Condensation Scrubbing of Fine Particles.
A.P.T., Inc., EPA Contract No. 68-02-1082. August, 1975.
5. Calvert, S., and S. Yung. Study of Horizontal Spray
Flux Force/Condensation Scrubber. A.P.T., Inc., EPA
Contract No. 68-02-1328, Task No. 10. July, 1976.
6. Calvert. S., S. Yung, and J. Leung. Entrainment Separators
for Scrubbers, Final Report, A.P.T., Inc. EPA Contract No.
68-02-0637. August, 1975.
7. Calvert, S., H.F. Barbarika, and S. Yung. Development
of Superior Entrainment Separators. A.P.T., Inc., EPA
Contract No. 68-02-2184. First Quarterly Progress
Report. December, 1976. (Unpublished).
8. Handbook of Emissions, Effluents and Control Practices
for Stationary Particulate Pollution Source. Midwest
Research Institute. Report to NAPCA. Contract No.
CPA 22-69-104. 1970
9. Peters, M.S., and K.D. Timmerhaus. Plant Design and
Economics for Chemical Engineers. New York. McGraw
Hill. 1968.
10. Taheri, M., and S. Calvert. Removal of Small Particles
from Air by Foam in a Sieve-Plate Column. J. Air
Pollution Control Association. 18: 240-245, 1968.
11. Whitmore, P.J.. Diffusiophoretic Under Turbulent Conditions,
Ph.D. Thesis, University of British Columbia, 1976.
133
-------
APPENDIX A
PROCESS DATA
134
-------
TABLE A-l. PROCESS DATA
F/C SCRUBBER DEMONSTRATION
Run
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
q'
K/R
0.15
0.14
0.13
0.12
0.10
0.12
0.14
0.19
0.24
0.11
0.24
0.13
0.16
0.22
0.18
0.13
0.18
0.28
0.29
0.28
Gas Flow
Am3/ sec
Gz
0.76
0.80
0.91
0.80
0.80
0.78
0.80
0.55
0.57
0.87
0.87
0.87
0.87
0.90
0.82
0.80
0.80
0.82
0.87
0.88
Gas Temperature
Tz
600
592
548
592
598
493
573
570
600
600
715
725
680
727
725
760
780
792
743
770
C
3
65
65
58
61
58
60
62
66
69
81
95
83
79
87
85
88
93
86
85
87
U
31
27
22
28
28
26
23
18
20
25
25
31
31
32
27
31
33
32
33
36
Liquid Flow Rate
4,'sec
Li
1.1
1.2
1.2
1.1
1.1
1.1
0.95
0.82
0.82
0.70
0.67
1.0
0.95
0.90
0.85
0.85
0.83
0.83
0.83
0.83
La
8.7
8.8
8.8
8.8
8.8
8.8
8.8
8.8
8.8
8.2
8.8
8.8
8.8
8.8
8.8
8.8
8.8
8.8
L,
8.6
8.8
8.8
8.8
8.8
8.8
8.8
8.8
8.8
8.2
8.1
7.8
7.8
7.4
7.1
7.1
7.2
7.0
6.9
6.9
!„ '
8.6
8.8
8.8
8.8
8.8
8.8
8.8
8.8
8.8
8.2
8.8
8.8
8.8
8,8
7.9
7.9
7.7
7.7
7.1
7.1
I.-
1.1
1.1
1.1
1.1
1.1
0.57
0.57
0.57
0.57
0.57
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
Liquid Temp.
— — *c —
T27
34
25
26
29
29
27
29
24
27
35
35
44
44
47
39
46
47
47
46
47
T2e
26
22
20
23
23
23
22
18
20
23
35
30
30
31
26
30
29
33
28
30
* See Figure 5-1 for explanation of symbols.
Continued
TABLE A-l (continued)
Run
No.
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
q'
K/E
0.28
0.28
0.26
0.20
0.32
0.32
0.38
0.35
0.38
0.30
0.28
0.40
0.35
0.27
0.13
0.29
0.31
0.24
0.30
0.24
Gas Flow
Am"/sec
G2
0.82
0.82
0.75
0.75
0.75
0.75
0.75
0.73
0.73
0.73
0.80
0.75
0.75
0.74
0.75
0.74
0.74
0.74
0.75
0.74
Gas Temperature
°C
Tz
725
732
717
730
780
802
825
795
820
696
710
775
760
443
466
417
417
340
510
450
T3
89
83
83
83
77
79
79
77
79
75
74
80
78
71
65
70
65
61
75
71
'U
39
40
40
41
43
36
35
34
33
34
33
36
33
37
34
37
35
36
38
35
Liquid Flow Rate
i.
L,
0.82
0.82
0.67
0.67
1.07
1.0
1.0
1.1
1.0
1.0
0.98
1.1
1.0
1.3
1.7
2.1
1.9
1.9
1.6
I,
8.8
8.8
8.8
8.8
8.8
8.8
8.8
8.8
8.8
8.8
8.8
8.8
8.8
8.4
9.3
8.8
8.8
8.9
9.2
9.2
/sec
La
6.6
6.6
6.3
6.3
6.7
6.2
6.2
5.8
5.8
5.8
5.9
5.9
5.6
7.1
7.8
7.6
7.3
7.6
7.6
8.2
L,,
7.2
7.0
7.1
7.1
7.6
6.7
6.7
6.3
6.3
7.0
7,0
6.4
6.4
7.2
8.2
7.8
7.7
8.0
7.8
8.6
L5
1.0
1.1
1.0
1.0
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
Liquid Temp.
°C
T27
36
36
40
40
42
47
59
51
52
43
42
49
48
45
46
44
46
44
47
42
T2a
25
28
26
28
30
32
34
33
32
30
28
33
34
33
34
34
38
35
38
34
Continued
135
-------
TABLE A-l (continued)
Run
No.
41
42
43
44*
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
— e/g
0.21
0.22
0.18
0.25
0.24
0.23
0.22
0.21-
0.20
0.21
0.30
0.22
0.22
0.25
0.31
0.21
0.22
0.28
0.22
Gas Flow
AmVsec
G,
0.74
0.73
0.81
0.81
O.E1
0.81
0.81
0.77
0.81
0.93
0.92
0.96
0.93
0.73
0.70
0.82
0.87
0.73
0.82
Gas Temperature
°c I
T,/T?
328
360
590
665
670
670
675
650
682
620
730
690
665
698
698
715
710
790
680
Th
68
66
67
73
72
71
70
69
68
70
76
71
69
71
75
70
70
74
70
T»
33
27
32
35
35
37
33
34
32
37
39
37
36
33
36
39
36
40
36
Liquid Flow Rate
J,/sec
— Ll 1
1.6
1.7
1.7
2.5
1.7
1.7
1.7
1.7
1.7
1.7
1.6
1.7
1.7
1.6
1.6
1.6
1.5
1.5
1.6
1.6
L2
9.1
8.8
8.9
9.1
9.3
9.2
9.2
9.2
9.1
9.1
8.9
8.7
8.6
8.8
8.8
8.8
8.9
8.8
8.8
8.9
^LT^
8.2
8.1
8.1
9.8
9.1
8.6
8.4
9.1
9.9
9.4
9.6
9.6
9.1
8.6
9.5
9.3
9.3
9.2
9.2
L»
8.3
7.8
7.9
10.7
9.3
9.1
8.6
9.8
10.1
9.9
9.8
9.9
9.3
9.3
9.1
9.6
9.5
9.5
9.5
9.5
Ls
1.1
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
Liquid Temp.
~ T2 7 1
45
40
39
37
40
45
45
40
41
36
48
52
49
48
48
46
45
38
52
43
IJB
36
32
30
33
32
35
35
33
31
31
36
37
36
36
33
37
40
30
39
38
*Ti value given instead of T2 from run 44 on.
Continued
TABLE A-l lcontinued)
Run
No.
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
q'
H/g
0.18
0.24
0.28
0.24
0.31
0.32
0.28
0.24
0.29
0.28
0.24
0.28
0.29
0. 24
0.19
0.22
0.19
0.24
Gas Flow
AmVsec
G8 .
0.82
0.82
0.82
0.80
O.S2
0.80
0.82
0.82
0.82
0.75
O.S2
O.E2
0.88
0.88
0.88
0.82
0.88
0.68
Gas Temperature
"C
Ti
685
710
725
710
770
750
730
690
735
700
725
725
740
661
565
635
520
600
T3
67
72
74
72
76
76
74
72
75
76
75
75
75
73
69
71
68
72
•u
34
39
36
38
42
39
38
38
41
50
38
38
38
36
38
37
36
38
Liquid Flow Rate
a
L!
1.6
1.6
1.6
l.S
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
L2
8.8
8.8
8.8
9.1
8.8
8.8
8.8
8.8
8.9
9.1
9.1
9.1
9.1
9.1
8.9
8.9
9.1
8.9
/sec
L3
9.2
9.2
9.2
9.2
9.2
0.2
9.1
9.1
9.1
9.1
9.1
9.1
9.0
8.9
9.0
8.9
9.0
8.9
L.
9.5
9.1
9.1
9.3
9.3
9.3
9.3
9.3
9.3
9.2
9. -2
9.2
9.1
9.1
9.1
9.2
9.1
9.1
Ls
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
1.3
Liquid Temp.
°C
T27
43
47
47
45
52
50
46
46
51
52
50
50
45
48
48
46
48
T28
38
38
35
35
41
39
34
37
39
29
38
35
37
34
37
36
35
36
136
-------
APPENDIX B
PARTICLE SIZE DATA
137
-------
Table B-l. INLET $ OUTLET SAMPLE PARTICLE DATA FOR RUN #3
IMP ACTOR
STAGE
NUMBER
Precutter
1
2
3
4
5
6
7
Filter
Sample
Volume
(DNm3)
INLET
cum
(mg/DNm3)
89.0
89.0
74.9
72.6
70.3
65.6
60.9
53.9
37.5
d
pc
(lamA)
6.4
15.8
11.7
7.23
2.94
1.47
1.04
0.66
0.080
OUTLET
M
cum
Og/DNm3)
23.9
22.3
20.8
20.8
20.8
20.8
20.0
18.5
11.5
V
OmA)
6.0
17.2
12.7
7.83
2.83
1.47
0.77
0.49
0.131
138
-------
Table B-2. INLET § OUTLET SAMPLE PARTICLE DATA FOR RUN #4
(SCRUBBER ONLY)
IMPACTOR
STAGE
NUMBER
Precutter
1
2
3
4
5
6
7
Filter
Sample
Vo lume
(DNm3)
INLET
M
cum
(mg/DNm3)
1910.0
1770.0
1760.0
1750.0
1740.0
1670.0
1130.0
498.0
257.0
d
pc
CvmA)
6.4
15.9
11.7
7.25
2.62
1.36
0.71
0.46
0.082
OUTLET
M
cum
(nig/ DNm3)
338.0
323.0
323.0
321.0
320.0
318.0
301.0
161.0
57.9
d
pc
(pmA)
6.0
17.0
12.5
7.74
2.80
1.46
0.76
0.49
0.129
Table B-3. INLET § OUTLET SAMPLE PARTICLE DATA FOR RUN #7
(SCRUBBER ONLY)
TMD A PTHD
JLMJrAL I UK
STAGE
NUMBER
Precutter
1
2
X
4
5
6
7
Filter
Sample
Volume
(DNm3)
INLET
Mcum
(mg/DNm3)
116.0
34.9'
34.9
33.2
33.2
29.9
28.3
16.6
So
(umA)
6.4
15.4
11.4
7.04
2.54
1.32
0.69
0.44
0.059
OUTLET
Mcum
(mg/DNm3)
33.7
32.9
32.9
32.2
31.4
29.2
27.7
18.0
V
(ymA)
6.0
15. 3
11.3
7.00
2.53
1.32
0.69
0.44
0.108
Table B-4. INLET § OUTLET SAMPLE PARTICLE DATA FOR RUN #42
(SCRUBBER ONLY!
IMPACTOR
STAGE
NUMBER
Precutter
1
2
3
4
5
6
7
Filter
Sample
Volume
(DNm3)
INLET
M
cum
(mg/DNm3)
395.0
393.0
341.0
340.0
336.0
328.0
309.0
224.0
94.0
d
PC
(pmA)
5.8
14.3
10.6
6.52
2.36
1.23
0.64
0.41
0.088
OUTLET
M
cum
(mg/DNm3)
121.0
120.0
120.0
116.0
113.0
112.0
107.0
72.4
27.4
d
pc
(pmA)
6.0
14.9
11.0
6.82
2.47
1.28
0.67
0.43
0.160
Table B-5. INLET B, OUTLET SAMPLE PARTICLE DATA FOR RUN #43
(SCRUBBER ONLY)
IMPACTOR
STAGE
NUMBER
Precutter
1
2
3
4
5
6
7
Filter
Sample
Volume
(DNm1)
INLET
Mcum
(mg/DNm3)
274.0
273.0
240.0
235.0
229.0
221.0
203.0
153.0
51.0
V
OmA)
5.8
14.1
10.4
6.44
2.33
1.21
0.63
0.41
0.086
OUTLET
Mcum
(mg/DNm3)
106.0
105.0
101.0
97.0
93.3
89.5
82.6
57.6
19.4
V
CvmA)
6.0
15.0
11.1
6.86
2.48
1.29
0.67
0.43
0.160
-------
Table B-6. INLET § OUTLET SAMPLE PARTICLE DATA FOR RUN #56
IMPACTOR
STAGE
NUMBER
Precutter
1
2
3
4
S
6
7
Filter
Sample
Volume
(DNm3)
INLET
M
cum
(mg/DNm3)
1170.0
647.0
618,0
577.0
515.0
368.0
294.0
• 274.0
141.0
d
PC
(ymA)
5.3
12.9
9.53
5.88
2.39
1.20
0.85
0.53
0.066
OUTLET
M
cum
[mg/DNm3)
48.4
40.1
40.1
39.5
38.8
36.3
33.7
28.6
15.3
,1
v
(VimA)
6.0
150
11 1
6 86
2.48
1.29
0.67
0 43
0.1S7
Table B-7. INLET § OUTLET SAMPLE PARTICLE DATA FOR RUN #58
IMPACTOR
STAGE
NUMBER
Precutter
i
•}
4
e
7
Filter
Sample
Volume
(DNm3)
INLET
"cum
(mg/DNms)
829.0
802.0
688.0
529.0
472.0
263.0
J92.0
157.0
124.0
dpc
(ymA)
6.7
16.5
12.2
7.54
3.06
1.54
1.09
0.68
0.085
OUTLET
Mcum
(mg/DNm3)
30.2
30.2
30.2
30.2
30.2
30.2
29.8
27.0
16.5
V
(umA)
6.2
15.5
11.4
7.06
2.55
1.33
0.69
0.44
0.249
Table B-8. INLET § OUTLET SAMPLE PARTICLE DATA FOR RUN »59
IMPACTOR
STAGE
NUMBER
Precutter
1
2
3
4
5
6
7
Filter
Sample
Volume
(DNm3)
INLET
Mcum
(mg/DNm3)
1250.0
390.0
375.0
369.0
342.0
211.0
145.0
122.0
98.6
V
(ymA)
6.1
15.4
11.4
7.02
2.85
1.43
1.01
0.64
0.052
OUTLET
Mcum
(mg/DNm3)
31.2
30.8
30.0
30.0
29.6
29.2
28.0
25.7
20.1
V
(ymA)
6.0
15.5
11.5
7.08
2.56
1.33
0.69
0.45
0.253
Table B-9. INLET $ OUTLET SAMPLE PARTICLE DATA FOR RUN #61
IMPACTOR
STAGE
NUMBER
Precutter
1
2
3
4
5
6
7
Filter
Sample
Volume
(DNm3)
INLET
Mcum
(mg/DNm3)
886.0
306,0
277.0
273.0
252.0
122.0
79.1
58.2
38.8
d
pc
GimA)
5.1
12.9
9.57
5.91
2.40
1.20
0.85
0.54
0.067
OUTLET
Mcum
(mg/DNm3)
53.3
53.3
53.3
53.3
53.3
53.0
52.6
48.0
31.2
V
CvmA)
6.0
15.5
11.5
7.10
2.56
1.34
0.69
0.45
0.262
-------
Tahlo B-10. INLET § OUTLET SAMPLE PARTICLE DATA FOR RUN #62
IMPACTOR
STAGE
NUMBER
Precutter
1
2
3
4
5
6
7
Filter
Sample
Vo lume
(DNm3)
INLET
M
cum
(rag/ DNm3)
] 150.0
505.0
463.0
446.0
395.0
217.0
137.0
103.0
69.2
d
pc
(ymA)
5.3
13.0
9.62
S.94
2.15
1.12
0.58
0.37
0.065
OUTLET
M
cum
(mg/DNm3)
51.7
32.4
32.0
32.0
31.6
30.8
29.2
25.3
16.2
d
V
(ymA)
6.0
15.4
11.4
7.04
2.54
1.32
0.69
0.44
0.253
Table B-ll. INLET § OUTLET SAMPLE PARTICLE DATA FOR RUN #64
TMP&rTHR
STAGE
NUMBER
Precutter
1
2
3
4
5
6
7
Filter
Sample
Vo lume
(DNm1)
INLET
Mcum
(mg/DNm3)
2490.0
953.0
918.0
900. 0
872.0
770.0
684.0
429.0
106.0
d
pc
(ymA)
5.2
12.9
9.51
5.87
2.12
1.10
0.57
0.37
0.064
OUTLET
Mcum
(mg/DNm3)
140. n
129.0
129.0
128.0
128.0
128.0
121.0
88.2
41.5
d
PC
(ymA)
6.1
15.6
11.5
7.11
2.57
1.34
0.70
0.45
0.250
Table B-12. INLHT 5 OUTLET SAMPLE PARTICLE DATA FOR RUN #66
IMPACTOR
STAGE
NUMBER
Precutter
1
2
3
4
5
6
7
Filter
Sample
Volume
(DNm3)
INLET
M
cum
(mg/DNm3)
1 23n .0
588.0
561 . 0
545.0
473.0
337.0
212.0
129.0
65.0
d
pc
(ymA)
5.2
12.9
9.51
5.87
2.12
1.10
0.57
0.37
0.063
OUTLET
M
cum
(mg/DNm3)
48.2
48.2
46.6
43.8
41.0
38.6
35.0
30.1
22.1
d
pc
(ymA)
6.1
15.5
11.4
7.06
2.55
1.33
0.69
0.45
0.249
Table B-13. INLET § OUTLET SAMPLE PARTICLE DATA FOR RUN #69
IMPACTOR
STAGE
NUMBER
Precutter
1
2
3
4
5
6
7
Filter
Sample
Volume
(DNm3)
INLET
Mcum
(mg/DNm3)
701.0
399.0
386.0
380.0
354.0
260.0
181.0
157.0
52.0
V
(ymA)
5.3
13.1
9.66
5.96
2.15
1.12
0.58
0.38
0.062
OUTLET
Mcum
(mg/DNm3)
77.1
75.1
74.7
74.7
74.2
73.4
71.8
61.7
35.3
V
(ymA)
6.1
15.4
11.4
7.04
2.5
1.3
0.69
0.44
0.246
-------
Table B-14. INLET § OUTLET SAMPLE PARTICLE DATA FOR RUN #71
IMPACTOR
STAGE
NUMBER
Precutter
1
2
3
4
5
6
7
Filter
Sample
Volume
(DNm3)
INLET
M
cum
(mg/DNm3)
1230.0
784.0
607.0
528.0
458.0
268.0
168.0
73.4
14.4
d
pc
(UmA)
5.3
12.9
9. 57
5.91
2.13
1.11
0.58
0.37
.063
OUTLET
M
cum
Cmg/DNm3)
d
pc
CumA)
t-o
'able B-15. INLET 8 OUTLET SAMPLE PART
TMDAfTnB
1 Mr AU 1 U rv
STAGE
NUMBER
Precutter
1
2
3
4
e.
7
Filter
Sample
Volume
(DNmJ)
INLET
Mcum
(mg/DNm3)
969.0
786.0
698.0'
569.0
549.0
440.0
354.0
237.0
124,0
V
(umA)
5.2
12.9
9.51
5.87
2.12
1.10
0.57
0.37
0.064
ICLE DATA FOR RUN #72
OUTLET
Mcum
(mg/DNm3)
80.7
64.2
63.4
63.4
63.4
63.4
63.4
58. 3
33.5
V
(umA)
6.0
15.3
11.3
7.00
2.53
1.32
0.69
0.44
0.254
Table B-16. INLET § OUTLET SAMPLE PARTICLE DATA FOR RUN #73
IMPACTOR
STAGE
NUMBER
Precutter
1
2
3
4
5
6
7
Filter
Sample
Vo lume
(DNmJ)
INLET
Mcum
Cmg/DNm3)
linn.o
270.0
232.0
214.0
198.0
123.0
89.6
64.0
25.6
V
CumA)
5.4
13.4
9.93
6.13
2.21
1.15
0.60
0.39
0.063
OUTLET
Mcum
Cmg/DNm3)
36.7
36.7
35. 9
34.7
32.8
31.2
29.2
26.4
21.3
V
CvmA)
6.2
15.5
11.4
7.05
2.55
1 .33
0.69
0.44
0.253
Table B-17. INLET § OUTLET SAMPLE PARTICLE DATA FOR RUN *74
IMPACTOR
STAGE
NUMBER
Precutter
1
2
3
4
5
6
7
Filter
Sample
Volume
(DNm5)
INLET
Mcum
(mg/DNm3)
1140.0
405.0
359.0
339.0
311.0
257.0
216.0
150.0
57.2
d
pc
(umA)
5.4
13.2
9.75
6.02
2.17
1.13
0.59
0.38
0.065
OUTLET
Mcum
(mg/DNm3)
49.0
48.6
48.2
47.4
45.8
45.8
44.7
40.0
24.5
dpc
(pmA)
6.1
15.5
11.5
7.09
2.56
1.33
0.69
0.45
0.253
-------
Table B-18. INLET § OUTLET SAMPLE PARTICLE DATA FOR RUN #75
IMPACTOR
STAGE
NUMBER
Precutter
1
2
3
4
5
6
7
Filter
Sample
Vo lume
CDNm3)
INLET
M „„
cum
Cmg/DNm3)
1200.0
434.0
398.0
375.0
345.0
246.0
169.0
112.0
68.4
d
pc
CumA.)
5.4
13.5
10.0
6.17
2.23
1.16
0.60
0.39
0.064
OUTLET
M
cum
Cmg/DNm3)
48.3
33.4
32.6
32.6
32.2
31.4
30.6
24.3
14.9
d
V
CymA)
6.1
15.4
11.4
7.03
2.54
1.32
0.69
0.44
0.255
Table B-19. INLET § OUTLET SAMPLE PARTICLE DATA FOR RUN #76
IMPACTOR
STAGE
NUMBER
Precutter
1
2
3
4
5
6
7
Filter
Sample
Volume
CDNm1)
INLET
Mcum
Cmg/DNm3)
2240.0
575.0.
413.0
378.0
334.0
198.0
143.0
97.3
70.6
d
PC
CuinA)
5.5
13.5
9.95
6.14
2.22
1.16
0.60
0.39
0.064
OUTLET
Mcum
Cmg/DNm3)
36.0
31.7
31.3
30.9
30.5
30.1
29.3
25.7
18.6
dpc
CpmA)
6.0
15.4
11.4
7.05
2.55
1.32
0.69
0.44
0.253
Table B-20. INLET § OUTLET SAMPLE PARTICLE DATA FOR RUN #77
IMPACTOR
STAGE
NUMBER
Precutter
1
2
3
4
5
6
7
Filter
Sample
Volume
(DNmJ)
INLET
M
cum
(mg/DNm3)
763.0
276.0
265.0
253.0
237.0
173.0
126.0
92.1
68.7
d
pc
CumA)
5.4
13.4
9.91
6.12
2.21
1.15
0.60
0.30
0.064
OUTLET
M
cum
(mg/DNm3)
SO. 5
33.1
3Q.8
30.0
29.6
28.8
28.4
24.9
15.4
d
PC
(pmA)
6.0
15.4
11.4
7.04
2.54
1.32
0.69
0.44
0.253
Table B-21. INLET § OUTLET SAMPLE PARTICLE DATA FOR RUN #78
IMPACTOR
STAGE
NUMBER
Precutter
1
2
3
4
5
6
7
Filter
Sample
Volume
CDNm3)
INLET
Mcum
(mg/DNm3)
1220 .0
459.0
407.0
382.0
331.0
144.0
58.0
32.9
18.8
dpc
CymA)
5.8
14.3
10.6
6.52
2.36
1.23
0.64
0.41
0.064
OUTLET
Mcum
Cmg/DNm3)
13.6
7.6
6.8
6.4
6.0
6.0
6.0
5.6
5.2
V
(pmA)
6.1
15.4
11.4
7.05
2.55
1.33
0.69
0.44
0. 251
-------
APPENDIX C
PARTICLE SIZE DISTRIBUTION PLOTS
144
-------
10.0
0.
10 20 30 40 50 60 70 80 90 95 98
MASS PERCENT UNDERSIZE
Figure C-l. Inlet and outlet size distributions
for run 3.
10.0
0.2
5 10
20 30 40 50 60 70 80 90 95 98
MASS PERCENT UNDERSIZE
Figure C-2. Inlet and outlet size distribution
for run 4 (scrubber only).
-------
10.0
= E RUN 7
li INLET
11 OUTLET
0.2
• ••••IIIIIIIIIIBMMMIMHIIIIIHIIUHIMn •(•••
5 10 20 40 60 80 90 95 98
MASS PERCENT UNDERSIZE
Figure C-3. Inlet and outlet size distributions
for run 7.
10
0.2
10 20 30 40 50 60 70 80 90 95 98
MASS PERCENT UNDERSIZE
Figure C-4. Inlet and outlet size distribution
for run 42 (scrubber only).
-------
10.0
RUN 43
INLET /\
OUTLET
0.2
10
20 30 40 50 60 70 80 90 95 98
MASS PERCENT UNDERSIZE
Figure C-5. Inlet and outlet size distributions
for run 43 (scrubber only).
10.
3H RUN 56
INLET
OUTLET
0.2
10
20 30 40 50 60 70 80 90 95 98
MASS PERCENT UNDERSIZE
Figure C-6. Inlet and outlet size distributions
for run 56.
-------
10.0
00
0.3
0.2
••IIIIIIII ••••• ••••! mil !•••••>•>• ••••• IIIIIHIIimBB
10 20 30 40 50 60 70 80 90 95 98 99 99.5
MASS PERCENT UNDERSIZE
Figure C-7. Inlet and outlet size distributions
for run 58.
10.0
0.4 i=f= = = = l
0.3
0.2
2 5 10 20 30 40 50 60 70 80 90 95 98
MASS PERCENT UNDERSIZE
Figure C-8. Inlet and outlet size distributions
for run 59.
-------
10.0
10.0
0.2
2 5 10 20 30 40 50 60 70 80 90 95 98 99 99.5
MASS PERCENT UNDERSIZE
0.2
5 10 20 30 40 50 60 70 80 90 95 98
MASS PERCENT UNDERSIZE
Figure C-9. Inlet and outlet size distribution for
run 61.
Fi-gure C-10. Inlet and outlet size distributions
for run 62.
-------
10.0
0.2
2 5 10 20 30 40 50 60 70 80 90 95 98
MASS PERCENT UNDERSIZE
Figure C-ll. Inlet and outlet size distributions
for run 64.
0.2
5 10 20 40 60 80
MASS PERCENT UNDERSIZE
90 95
98
Figure C-12. Inlet and outlet size distribution
for run 66.
-------
10.0
0.2
5 10 20 40 60 80 90 95 98
MASS PERCENT UNDERSIZE
Figure C-13. Inlet and outlet size distribution
for run 69.
10.0
ft
0.2
0.5
5 10 20 40 60
MASS PERCENT UNDERSIZE
80 90
Figure C-14. Inlet size distribution for run 71.
-------
10.0
cn
oj
p.
0.3
0.2
10 20 40 60 80 90 95 98
MASS PERCENT UNDERSIZE
Figure C-15. Inlet and outlet size distribution
for run 72.
10.0
5.0
4.0
3.0
2.0
nj
P.
1.0
0.5
0.4
0.3
0.2
2 5 10 20 40 60 80
MASS PERCENT UNDERSIZE
90 95
Figure C-16. Inlet and outlet size distribution
for run 73.
-------
10.0
0.2
10
20 30 40 50 60 70 80 90 95 98
MASS PERCENT UNDERSIZE
Figure C-17. Inlet and outlet size distribution
for run 74.
10.0
0.2
10 20 30 40 50 60 70 80 90 95 98
MASS PERCENT UNDERSIZE
Figure C-18. Inlet and outlet size distributions
for run 75.
-------
10.0
0.2
10 20 30 40 SO 60 70 80 90 95 98
MASS PERCENT UNDERSIZE
Figure C-19. Inlet and outlet size distributions
for run 76.
10.0
= RUN 77
= INLET
BOUTLET
0.3
0.2
5 10 20 40 60 80
MASS PERCENT UNDERSIZE
90 95 98
Figure C-20. Inlet and outlet size distribution
for run 77.
-------
10.0
rt
P<
0.5 1
5 10 20 40 60
MASS PERCENT UNDERSIZE
80
90
Figure C-21. Inlet size distribution for run 78.
155
-------
APPENDIX D
IMPINGER DATA
156
-------
TABLE D-l. IMPINGER DATA AND RESULTS
TABLE D-l. (continued)
Ol
Run
No.
1
2
3*
4*
5
6
7*
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
Quencher Inlet
Concentration
mg/DNm3
110.0
132.0
1,580.0
N.A.
II
II
II
II
II
II
11
II
11
11
II
II
111.0
62.1
45.2
102.0
179.0
172.0
ci-
io, ooo
10,000
1,100
2,900
450
380
120
1,600
2,000
pH
1.7
1.3
2.3
1.9
2.5
2.4
2.5
2.1
1.9
Scrubber Outlet
Concentration
mg/DNm3
85.2
19.1
267.7
8.0
24.1
14.9
17.7
2.1
2.9
0.6
1.0
13.8
18.8
12.7
11.4
18.1
0
14.9
5.1
12.4
2.1
0.7
ci-
120
300
45
50
8,000
< 1
< 1
< 1
< 1
< 1
< 1
39
32
< 1
30
39
80
170
20
13
50
78
PH
5.6
13.3
6.8
5.7
2.4
4.0
4.0
4.2
4.1
3.3
3.3
3.2
3.3
4.2
3.3
3.3
3.3
3.5
5.5
4.0
3.7
3.3
Pt
1
77.5
14.5
16.9
0
24.0
11.3
12.1
1.2
0.4
Run
No.
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42*
43*
44
Quencher Inlet
Concentration
mg/DNm3
2(17.0
161.0
199.0
103.0
125.0
38.0
18.4
19.6
15.5
38.0
17.8
0
0
40.0
12.7
12.0
0
38.3
11.2
N.A.
II
154.5
Cl~
1,800
10,000
1,100
220
900
1,400
20
32
80
78
34
250
20
90
20
400
200
33
39
500
pH
1.9
1.3
2.3
2.6
2.2
2.2
4.0
3.1
3.0
2.7
2.8
2.6
3.1
3.1
3.3
2.8
2.6
2.8
2.9
2.4
Scrubber Outlet
Concentration
mg/DNm s
1.3
1.3
0
0
0
5.6
7.1
10.1
14.0
7.1
9.1
0
0
28.0
9.2
11.9
0
0
0
0
0.4
38.4
Cl"
32
20
90
< 1
4
12
4
8
4
8
8
1
4
1
10
20
< 1
2
4
< 1
< 1
8
PH
3.2
3.3
3.3
4.4
4.1
4.9
4.9
6.4
4.2
4.5
5.1
4.9
4.4
4.7
3.8
4.1
4.2
4.4
4.7
4.8
4.6
3.9
Pt
%
0.6
0'.8
0
0
0
14.7
38.6
51.5
90 = 3
18.7
51.1
0
0
70.0
72.4
99.0
0
0
0
24.8
•Cascade impactor runs.
*Cascade impactor runs.
-------
TABLE D-l. (continued)
TABLE D-l. (continued)
tn
oo
Run
No.
45
46
47
48
49
50
51
52
53
54
55
56*
57
58*
59*
Quencher Inlet
Concentration
mg/DNm3
**
111.0
**
98.2
**
103.0
**
103.0
**
101.0
**
171.0
**
76.7
**
26.5
**
4.9
**
66.2
**
131.0
86.3
160.0
0
121.0
Cl
20
25
240
180
470
280
120
180
125
310
2,100
30
< 1
100
20
pH
2.9
2.9
2.5
2.8
2.3
2.4
2.9
2.5
2.8
2.3
1.9
3.2
4.8
2.4
3.0
Scrubber Outlet
Concentration
mg/DNm3
53.3
14.9
1.0
0
4.5
4.5
0
0
0
12.0
6.9
28.1
0
0
0
• ci-
1
1
< 1
55
14
20
13
8
18
15
9
21
< 1
9
< 1
pll
4.1
4.2
3.8
4.2
3.7
4.1
3.S
4.2
4.1
4.2
4.0
4.4
4.2
4.2
4.2
Pt
V
48.2
15.2
1.0
0
4.4
2.6
0
0
0
18.1
5.3
32.5
0
0
0
* Cascade impactor runs.
** Furnace outlet.
Run
No.
60
61*
62*
63
64*
65
66*
67
68
69*
70
71*
72*
73*
74*
75*
76*
77*
78*
Quencher Inlet
Concentration
mg/DNm3
* *
90.7
0
N.A.
**
30.0
281.0
ft *
10.3
8.1
0
**
10.7
325.8
N.A.
363.4
364.0
189.0
33.8
267.0
269.0
436.0
20.3
ci-
1,500
6
120
550
250
500
40
200
500
120
250
260
62
200
80
70
80
40
pH
1.8
3.7
2.7
2.2
2.3
2.7
3.3
2.6
2.3
2.6
2.4
2.2
2.6
2.3
2.3
2.4
2.4
2.8
Scrubber Outlet
Concentration
mg/DNm3
1.0
0
45.2
19.6
3.2
0
0
8.4
0
0
0
0
0
9.6
3.2
8.8
38.4
ci-
< 1
< 1
< 1
4
8
< 1
< 1
30
<: 1
2.5
20
10
< 1
10
< 1
< 1
< 1
< 1
pH
4.9
4.4
4.2
4.6
4.3
4.7
4.6
3.5
4.5
4.4
5.1
5.2
4.6
4.2
4.3
4.5
4.2
4.7
Pt
%
1.1
0
7.0
31.1
0
0
78.5
0
0
0
0
3.6
1.2
2.0
* Cascade impactor runs.
** Furnace outlet.
-------
APPENDIX E
CALCULATION OF PARTICLE NUMBER CONCENTRATION
FROM CASCADE IMPACTOR DATA
159
-------
APPENDIX E
CALCULATION OF PARTICLE NUMBER CONCENTRATION
FROM CASCADE IMPACTOR DATA
The F/C model for prediction of scrubber performance uses
particle number concentration as one of its parameters. Data
on particle size distribution by mass as taken with a cascade
impactor can be used to compute the particle number concentra-
tion. The procedure used in this study to calculate particle
count using size distribution data is summarized below.
Particle number concentration is related to mass concen-
tration and the mass mean diameter by equation (E-l).
where n = number concentration, number/DNcm3
c = mass concentration, g/DNcm3
p = particle density, g/cm3
d = mass mean diameter of particle, cm
The mass mean diameter is defined by:
E nj dj3
dm - * (B-2)
where "i" represents an individual particle size. If the
distribution of diameter is log normal over the whole range
of sizes, the relationships between various mean diameter are
simple. The mass mean diameter is related to the mass median
(geometric mass mean) diameter by:
lndm= 1"^ - I-* 1" a*
160
-------
where d - mass median diameter, cm
a = geometric standard deviation
o
In the demonstration program the particle size distri-
butions were measured gravimetrically using the cascade
impactor as a device to fractionate particles of different
size. Cumulative mass concentrations of particles collected
at each stage and all the stages below, including the absolute
filter, were calculated. The cut diameters for the impactor
stages were calcualted from the sampling rate in conjunction
with the calibration data for A.P.T. impactors. These data
are presented in Appendix "B" while the size distribution
plots are given in Appendix "C".
The particle size data obtained from cascade impactors are
reported in terms of aerodynamic diameter. Equation (E-4)
shows the relationship between aerodynamic and physical
diameters.
d = d (p C')1/2 fE-4}
pa p. p J ^
where d = aerodynamic diameter, ymA
pa
d = particle diameter, ym
p = particle density, g/cm3
C1 = Cunningham correction factor, dimensionless
Bimodal Size Distribution
The demonstration plant inlet data showed that the particles
entering the quencher were widely dispersed and possessed a
bimodal distribution. The particle population was a mixture
of very small condensation aerosol and large size particles
from the sodium carbonate spray in the cross-over duct to the
quencher unit. As operated the cascade impactor data covered
only the range of approximately 0.5 ymA to 6 ymA. The cut
point of the last stage of the impactor was about 0.5 ymA,
while the cut point of the pre-cutter was around 6 ymA.
Since the data over the whole size range of the particles
entering the quencher were not log normal, it was not possible
161
-------
to define a true mass median diameter and "a " simply from a
straight line plot of size distribution on log probability
paper. An alternate rationale for defining particle size
distribution in a way useful for computing number concentra-
tion had to be devised.
Using equations (E-l) to (E-4) it can be shown that the
majority of particles by number are in the submicron range
and therefore the size distribution in the larger diameter range
is unimportant for determining number concentration. Conse-
quently, one can compute number concentration from hypothetical
values of "d " and "a " obtained by assuming log normal
distribution over only the cumulative mass data below about
1 ymA. This was done in a consistent manner by plotting the
distribution line through only the last two data points and
taking "d " and "a " from that line.
Inlet "n"
The inlet size data deviated more from log normal than
the outlet and also the outlet data were more extensive in
the small particle size range than the inlet. It was found
that more consistent results were obtained by computing inlet
number concentration from outlet data than from inlet data.
The computation method used the outlet particle number concen-
tration in conjunction with the penetration curve for each run.
It was established that the majority of particles by number
(over 90% } were below 0.5 ymA and that the penetration
curve for the F/C scrubber in the region below this size was
constant due to particle growth characteristics, as shown in
Figure E-l. Inlet particle number concentration was determined
from the following:
n
n
.
in Pt*
where n Qut = calculated number concentration at outlet,
no. particles/DNcm3 using equation (E-l)
Pt* = constant penetration value for particles below
about 0.5 ymA size
162
-------
0.3
0.2
o
I—I
IH
0.1
o
I—I
H
H 0.05
w
S 0.04
0.03
PH
0.02
EXPERIMENTAL
EXTRAPOLATED
0.01
iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii inn inn •••••••••! IIIIIIIIIIIIIIIHIIIIMIIIIIH
0.3
0.5
1.0 1.5 2.0 3.0
d , ymA
pa.
Figure E-l. Particle penetration versus
aerodynamic diameter for
run 56.
163
-------
n - = calculated number concentration at inlet,
no. particles/DNcm3
The values of "n ^ " as determined from equation (E-5)
were used for predicting the performance of the F/C
demonstration system in conjunction with the methods presented
in Chapter 7. The validity of the use of a constant value
of "Pt" in the procedure was established by comparing number
concentrations computed by detailed accounting for "Pt" as a
function of particle size and the integration over the entire
size distribution with concentrations computed by the simpler
procedure. There was littlu difference between the two
methods. These values are given in Table E-l with the corre-
sponding "d " and "a " from hypothetical distribution based
on the last two stages.
Summary
In summary, the following procedure is followed for calcu-
lation of particle number concentration for each run:
1. Determine cumulative mass loadings for each stage of
cascade impactor. This stage data are presented in
Appendix "B".
2. Calculate and plot size distribution by plotting
percent of mass undersize for each stage versus
particle size.
3. Determine hypothetical values of "d " and "a " for
r O O
the outlet size distribution using only the last
two data points in the small particle region.
4. Calculate the outlet particle number concentration
using equations (E-l) and (E-3).
5. Calculate the inlet particle number concentration
using equation (E-5) in conjunction with the experi-
mental penetration curves as presented in Chapter 6
for each run.
164
-------
TABLE E-l. PARTICULATE DATA F/C SCRUBBER DEMONSTRATION
Run
No.
3
4**
42**
43**
56
58
59
61
62
64
66
69
72
73
74
75
76
77
Inlet Particulate*
Pt*,Fr.(a}
0.22
0.42
0.32
0.55
0.11
0.25
0.30
0.65
0.25
0.19
0.11
0.21
0.19
0.16
0.20
0.21
0.25
0.27
n x 10"7
no./DNcm3
8.2
3.6
4.3
1.7
19.0
4.4
67.0
3.8
40.0
13.0
380.0
13.0
16.0
30.0
9.0
31.0
52.0
14.0
Outlet Particulate*
VUmA
0.50
0.78
0.60
0.64
0.58
0.42
0.33
0.40
0.70
: 0.58
0.48
0.45
0.50
0.35
0.45
0.65
0.43
0.70
a
g
2.0
1.6
1.6
1.6
1.9
1.5
2.1
1.5
2.7
1.7
2.7
1.6
1.7
3.1
1.6
2.5
2.3
2.4
n x 10"7
no./DNcm3
1.8
1.5
1.4
0.94
1.5
1.1
20.0
2.5
9.9
2.5
42.0
2.8
3.1
356.0
1.8
6.4
13.0
3.7
* Particulate data across quencher inlet and scrubber inlet
unless otherwise noted.
** Particulate data across plate scrubber only.
NOTE: (a) Pt* is the experimental penetration over the small
particle range (less than about 0.5 ymA).
16 b
-------
EXAMPLE CALCULATION
For illustration purposes, an example calculation based on
the above procedure is given below. Data from run 56 are used
to calculate the particle number concentration.
Step 1: The stage data as collected by the cascade impactor
for run 56 are given in Table B-6. The mass percent undersize
is calculated for each stage based on the total mass collected
by the impactor. These values are presented in Table E-2.
Step 2: The size distribution (d versus mass percent under-
pc
size) is plotted on log probability paper as shown in Figure E-2.
Step 3: The "d " and "a " for the outlet size distribution are
Po &
approximated using only the data from the last two stages of
the impactor. This is represented by a dashed line in
Figure E-2. For run 56, the "d " and "a " on the outlet
Jr & o
were approximated as 0.58 ymA and 1.9.
Step 4: The "d " of the outlet size distribution is in terms
c— pg
of aerodynamic diameter. The physical diameter of an equiva-
lent sphere is calculated as 0.22 jam using equation (E-4) and
particle density of 4 g/cm3. The "
-------
TABLE E-2. INLET AND OUTLET SAMPLE DATA FOR RUN 56
Impactor
Stage
Number
Precutter
1
2
3
4
5
6
7
Filter
Inlet
cum
Mg/DNm3
1,170
647
618
577
515
368
294
274
141
% by Mass
Under
100.0
55.3
52.8
49.3
44.0
31.4
25.1
23.4
12.1
Sc
ymA
5.3
13.0
9.5
5.9
2.4
1.2
0.85
0.53
Outlet
Mcum
Mg/DNm3
48.4
40.1
40.1
39.5
38.8
36.3
33.7
28.6
15.3
% by Mass
Under
100.0
82.8
82.8
81.6
80.1
75.0
69.6
59.1
31.6
V
6.0
15.0
11.0
6.9
2.5
1.3
0.67
0.43
167
-------
10.0
ca
INLET
OUTLET O
Hypothetical log-normal
size distribution for
smaller particles
2 5 10 20 40 60 80
MASS PERCENT UNDERSIZE
Figure E-2, Inlet and outlet size distribution
for run 56.
95 98
168
-------
APPENDIX F
EXAMPLE CALCULATION
169
-------
APPENDIX F
EXAMPLE CALCULATION AND PREDICTION
OF FRACTIONAL AND OVERALL PENETRATION
SUMMARY
The sequence of steps to be followed in predicting the
performance of an F/C scrubber system involving a sieve plate
column are presented in detail in Chapter 7. In outline, the
prediction of the fractional and overall penetration of a flux
force scrubber is based on the following steps:
1. Determine the initial particle size distribution.
2. Compute particle penetration from the saturator.
3. Compute particle penetration due to inertial im-
paction on the first plate of the sieve plate
column.
4. Calculate the volume fraction of gas condensing,
"f ", and then calculate the penetration due to
diffusiophoresis.
5. Determine the grown particle size distribution.
6. Compute the particle penetration function due to
inertial impaction for the remaining stages of the
sieve plate scrubber.
7. Use the grown particle size distribution and the
particle penetration function of step 6 to compute
the penetration due to inertial impaction after growth.
8. Compute the penetration of the F/C scrubber as the
product of the penetrations due to steps 2, 3, 4
and 7.
DATA (Initial Conditions)
The data for run 56 will be used for illustration purposes.
1. Size distribution of the particulate matter at inlet,
see Table B-6.
170
-------
2. Particle number concentration, n = 1.3 x 108/DNcm3
(see Appendix "E" for calculation procedure).
3. Condensation ratio, q' = 0.31 g/g (see Appendix "A").
4. Inlet absolute humidity, H = 0.35 g/g
5. Particle density, p =4.0 g/cm3
6. Scrubber operating mode, I (see Table 6-3).
CALCULATION PROCEDURE
Step 1: The cumulative mass fraction undersize is tabulated as
a function of initial particle aerodynamic size, d , from the
pel j
inlet size data given in Table B-6 and plotted in Figure E-2.
Column 1 of Table F-l lists the cumulative mass fraction under-
size, while the corresponding particle diameter is given in column
2. The initial aerodynamic diameter, d , is converted to physical
diameter, d , using equation E-4. Column 3 gives the physical
particle diameter, d
F ' Pi
Step 2: The next step is to compute the particle penetration from
the saturator (Pt ) based on the saturator collection efficiency charac-
3.
teristics and the initial particle size distribution. Since the
saturator is not very efficient over the submicron particle size
range of less than 1.5 ymA and the main purpose of this calcu-
lation is to determine the penetration curve over this size range,
the particle penetration from the saturator was neglected.
Step 3: The collection efficiency relationship for sieve plate
as given in equation 7-8 was utilized to compute the particle
penetration due to inertial impaction during the bubble formation
on the first plate (pt>J • Figure F-l presents the penetration curve com-
puted from this relationship. The first plate impaction pene-
tration for each size can be read from curve no. 1 in Figure
F-l and the results are listed in column 5 of Table F-l.
Curves 2, 3, and 4 are for the total penetration for the
sequence of plates from 2 through 6. A constant gas flow of
50 Am3/min and foam density of 0.4 is assumed.
At this point it should be noted that data reduction of
all the runs in the main text (Chapter 6) were based on the
171
-------
simplifying and optimistic assumption that growth had occurred
before the first plate. Therefore, predicted penetrations for
the runs in the main text are slightly lower than actually the
case using the more conservative performance prediction proce-
dure followed in this appendix, which assumes insoluble
particles and no particle growth before the first plate.
Step 4: The next step is to determine the penetration for col-
lection by diffusiophoresis, Ptc. Equation 7-10 is utilized
to calculate "Ptc". It will assume that the vapor condensing
on the particles is completely utilized in causing agglomeration.
Since the inertial impaction efficiency is sufficiently high,
most of the particles swept to other particles will eventually
be collected by impaction. In order to calculate Ptc, the
volume fraction of gas condensing, f , is calculated from the
condensation ratio, q1. For "q"' equal to 0.31 g/g, the volume
fraction of gas condensing is calculated to be 0.32. "Pt " is
calculated from equation 7-10:
Ptc = l-0.85(fv) = 0.73
The diffusiophoretic penetration applies equally to all particle
sizes as listed in column 6 of Table F-l.
Step 5: This step determines the grown particle size leaving the
condenser (sieve plate column) from the values of "q" and "f ".
Figure F-2 is a size distribution plot showing lines for the par-
ticles before and after particle growth. The conditions used for
this plot were:
n = 1.3 x 108/DNcm3 @ 0°C
q' = 0.31 g/g
f = 0.25 (see Chapter 7)
Column 3 lists the grown particle diameter, d , for each d
pa2 pi'
Step 6: The particle penetration (Pt^) due to inertial impaction
for the remaining plates 2-6 is calculated. The collection effi-
ciency relationship for sieve plate is given in equation 7-8, and
172
-------
the results for different operating configurations at the demon-
stration plant are presented in Figure F-l.
Step 7: The relationship between Pt.N and d (from step 6) and
i pa r
the grown size distribution (from step 5) are used to compute
the fractional penetration due to inertial impaction after growth
(Pt,). Column 7 of Table F-l lists these values for each d • .
a pa2
Step 8: The total fraction penetration for the F/C scrubber,
Ptt, is calculated as follows and presented in column 8 of Table
F-l:
Ptt = Pta x Ptb x Ptc x Ptd
where "pta" ^ue to impaction in the saturator (neglected)
"Ptb" due to impaction in the condenser (column 5)
"Ptc" due to diffusiophoresis in the condenser (column 6)
"Pt^" due to impaction in stages after the condenser
(column 7)
The total fraction penetration, Pt , is plotted for each initial
particle size, d , to give the predicted penetration curve (see
paj
Figure F-3) .
In addition, the overall penetration can be determined by inte-
grating the fractional penetration, Ptt (column 8), over the
entire range of initial size distribution (column 1). This inte-
gration can be accomplished graphically (see Figure F-4) or alge-
braically. Using trapezoidal approximation, the integration is
done algebraically using a programmable calculator to give:
pT = 0.043 = 4.3%
Thus, the predicted overall penetration of the F/C scrubber at
given initial conditions is 4.3%.
173
-------
TABLE F-l. EXAMPLE CALCULATION FOR PREDICTION OF FRACTIONAL AND
OVERALL PENETRATION OF F/C SCRUBBER SYSTEM (RUN 56, configuration I)
Initial Particulate
Mass Fr.
1
0
0.10
0.21
0.25
0.27
0.29
0.30
0.30
0.31
pai
ymA
2
0
0.50
0.75
1.0
1.25
1.50
1.75
2.0
2.25
d
Pi
ym
3
0
0.18
0.30
0.43
0.55
0.67
0.80
0.92
1.05
Grown
pa2
yraA
4
1.22
1.22
1.25
1.32
1.43
1.59
1.77
2.0
2.25
Penetration, Fraction
Ptb
5
1.0
0.97
0.94
0.89
0.83
0.77
0.68
0.62
0.55
Ptc
6
0.73
0.73
0.73
0.73
0.73
0.73
0.73
0.73
0.73
Ptd
7
0.23
0.23
0.22
0.185
0.145
0.076
0.045
0.019
~
Ptt
8
0.17
0.16
0.15
0.12
0.088
0.043
0.022
0.0089
~
-------
10.0
E-H
O
tL.
o
Hi
H
w
•J
cj
i.o
0.5
0.4
0.3
0.2
0.1
0.05 !;
0.03
Plate 1 (all config
Plate 2-6 (A-B)
Plate 2-6 (D-H)
Plate 2-6 (C, 1-3)
Gas Flow = 50 AmVmin
F = 0.4
UG = 1.8 x Itf'g/cin-s
0.3 0.4 o.S
1.0
1.5 2.0 3.0
PARTICLE DIAMETER (dp&),
Figure F-l. Scrubber penetrations for
collection by inertial im-
pact ion as computed from
equation 7-8 for different'
operating configurations
(see Table 6-2).
Initial Conditions
1. n = 1.3 x 108DNcm3
0.2
5 10 20 30 40 50 60 70 80 90 95 98
MASS % UNDERSIZE
Figure F-2. Predicted grown particle size
distribution (!Um 56 data) .
-------
0.3
Experimental
Predicted
0.01
0.3 0.5 1.0 l.S 2.0 3.0
dpa, ymA
• Figure F-3. Particle penetration
versus aerodynamic
diameter fRun 56 data).
§
1-H
i
0.04 0.08 0.12 0.16 0.2 0.24 0.28
CUMULATIVE MASS UNDERSIZE, FRACTION
0.32
Figure F-4. Prediction of overall penetration for Run 56 using
graphical integration (Run 56 data).
-------
TECHNICAL REPORT DATA
(Please read fmtructions on the reverse before completing)
REPORT NO.
EPA-600/2-77-238
2.
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Fine Particle Collection by a Flux-Force/Condensa-
tion Scrubber: Pilot Demonstration
5. REPORT DATE
December 1977
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Seymour Calvert and Shamim Gandhi
8. PERFORMING ORGANIZATION REPORT NO
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Air Pollution Technology, Inc.
4901 Morena Boulevard, Suite 402
San Diego, California 92117
10. PROGRAM ELEMENT NO.
1AB012; ROAP 21ADL-002
11. CONTRACT/GRANT NO.
68-02-1869
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; 6/74-6/77
14. SPONSORING AGENCY CODE
EPA/600/13
is.SUPPLEMENTARY NOTES IERL-RTP project officer for this report is Dale L. Harmon,
Mail Drop 61, 919/541-2925.
IB. ABSTRACT Tne report gives results of B. pilot-scale demonstration of flux-force/conden-
sation (FF/C) scrubbing for fine particle control, carried out on a secondary metal
recovery furnace. Results were consistent with those of preceding laboratory bench-
scale and pilot-plant studies. The system was generally capable of 90-95% efficiency
on particles with amass median aerodynamic diameter of 0.7-0.8 micrometers A,
achieved with a 68 cm W.C. gas-phase pressure drop. A conventional high energy
scrubber without FF/C effects would require pressure drops of roughly 250 cm
W. C. for 90% and 535 cm W. C. for 95% particle collection efficiency. FF/C effects
are those which accompany the condensation of water vapor from the gas and are
generally caused by contacting hot humid gas with colder liquid and/or by injecting
steam into saturated gas. Mathematical models have been developed for predicting
FF/C effects and for use in scrubber system design. Agreement between the model
predictions and experimental results was good. The report gives FF/C system design
details, experimental results, analysis of results, description of mathematical
models, design of an optimized system, cost estimates, and recommendations for
future research.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
Air Pollution
Scrubbers
Condensing
Dust
Iron and Steel
Industry
Furnaces
Mathematical
Models
Air Pollution Control |13B
Stationary Sources 07A
Flux-Force/Condensa- K)7D
tion Scrubbers ftlG
Particulate
Secondary Metal Reco-
very f!3A
12A
3. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (ThisReport)
Unclassified
[21. NO. OF PAGES
195
20. SECURITY CLASS (Thispage)
Unclassified
|22. PRICE
EPA Form 2220-1 (9-73)
177
------- |