vxEPA
United States
Environmental Protection
Agency
Industrial Environmental Research
Laboratory
Research Triangle Park NC 2771 1
EPA-600/7-79-071
February 1979
Mobile Bed Flux
Force/Condensation
Scrubbers
Interagency
Energy/Environment
R&D Program Report
-------
RESEARCH REPORTING SERIES
Research reports of the-Qffice of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8, "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from the
effort funded under the 17-agency Federal Energy/Environment Research and
Development Program. These studies relate to EPA's mission to protect the public
health and welfare from adverse effects of pollutants associated with energy sys-
tems. The goal of the Program is to assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations include analy-
ses of the transport of energy-related pollutants and their health and ecological
effects; assessments of, and development of, control technologies for energy
systems; and integrated assessments of a wide'range of energy-related environ-
mental issues.
EPA REVIEW NOTICE
This report has been reviewed by the participating Federal Agencies, and approved
for publication. Approval does not signify that the contents necessarily reflect
the views and policies of the Government, 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 Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/7-79-071
February 1979
Mobile Bed Flux
Force/Condensation
Scrubbers
by
S.C. Yung, R. Chmielewski, and S. Calvert
Air Pollution Technology, Inc.
4901 Morena Boulevard, Suite 402
San Diego, California 92117
Contract No. 68-02-2124
Program Element No. EHE624A
EPA Project Officer: Dale L. Harmon
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
ABSTRACT
Fine particle collection in mobile bed scrubbers has been
determined experimentally. Particle collection efficiency
increased greatly as the gas phase pressure drop increased.
With no water vapor condensation, the performance capability of
a mobile bed scrubber is less than that of a gas-atomized spray
scrubber with the same pressure drop. Compared to packed bed
and sieve plate scrubbers , the mobile bed has better efficiency
when the pressure drop is above 20 cm W.C. The presence of
limestone in the scrubber liquid has no effect on particle
collection.
When the mobile bed scrubber was used as a flux force/
condensation scrubber it had better performance characteristics
than sieve plate and spray scrubbers with condensation.
However, its capability is inferior to a F/C system consisting
of a condenser and venturi scrubber.
None of the design equations reported in the literature
are adequate to predict the collection efficiency and pressure
drop of the mobile bed scrubber. New correlations have been
developed in this study to predict particle collection and
pressure drop.
Liquid entrainment characteristics have also been deter-
mined experimentally in this study. This information may
be used to design high efficiency entrainment separators for
mobile bed scrubbers.
This report was submitted in partial fulfillment of con-
tract number 68-02-2124, by Air Pollution Technology, Inc.
under the sponsorship of the U.S. Environmental Protection
Agency. This report covers the period from November 3, 1975
to November 3, 1978.
111
-------
CONTENTS
Page
Abstract iii
Figures vi
Tables xii
Abbreviations and Symbols xv
Acknowledgment xix
Sections
1. Summary and Conclusions ..... 1
Summary 1
Conclusions 16
2. Introduction 19
3. Preliminary Studies 22
Entrainment Measurements 22
4. Mobile Bed F/C Scrubber Pilot Plant 55
Mobile Bed F/C Scrubber Pilot Plant 55
Instrumentation and Calibration 67
Particle Generator 68
Particulate Sampling System 68
Data Reduction Methods 70
5. Experiments 75
Experimental Conditions Studied 75
Experimental Procedures 77
Experimental Results 79
Discussions 79
IV
-------
CONTENTS (continued)
Pat
6. Mathematical Modeling 122
Literature Search 122
Comparison of Experimental Data with
Predictions 142
Mathematical Modeling. . . ' 148
F/C Scrubbing 152
7. Evaluation of the Mobile Bed Scrubber 174
Particle Scrubbing 174
Mobile Bed as a F/C Scrubber 176
Potential for Power Plant Application 184
Mobile Bed Scrubber Design Recommendations . . . 188
8. Future Research Recommendations 191
References 198
Appendices
A DC-1 Drop Counter Entrainment Data 202
B Test Conditions and Particle Data 210
C Grade Penetration Curves for Cold Operation Mode. . 217
D Slurry Scrubbing Penetration Curves 235
E Grade Penetration Curves for F/C Runs 237
-------
FIGURES
Number Page
1 Cut/power relationship for scrubbers ......... 14
2 Mobile bed scrubber system for entrainment
measurement .................. .... 23
3 Entrainment measurement sampling train ........ 29
4 Comparison between cascade impactor and DC-1
drop counter data .................. 33
5 Sketch of experimental equipment ........... 35
6 "Dry salt" particle size distribution ......... 36
7 Predicted drop size distribution as a function
of water evaporated ................. 38
8 Measured wet size distributions, U. of W.
impactor and series glass impactor ......... .39
9 Experimental setup for glass impactors ........ 41
10 Wet size distribution obtained with
glass impactors ................... 42
11 Comparison between cascade impactor and
DC-1 drop counter data ................ 44
12 Mass median drop diameter of entrainment
from mobile bed scrubber (Drop counter data) ..... 49
13 Cumulative entrainment loading (Drop
counter data) .................... 50
14 Cumulative entrainment loading (Drop
counter data) .................... 51
15 Cumulative entrainment loading (Drop
counter data) ................... 52
16 Entrainment flow rate (Drop counter data) .... 54
VI
-------
FIGURES (continued)
Numb e r Page
17 Process flow diagram of mobile bed
F/C scrubber system 56
18 Mobile bed layout 59
19 Mobile bed and entrainment separator 60
20 Cooling tower framing and mobile
bed sumps 61
21 Quencher, furnace and blower vibration mounts. ... 62
22 Mobile bed platform and cooling tower
piping elevation 63
23 Mobile bed scrubber 64
24 Powder redispersion particle generator 69
25 Pressure drop across one stage of a mobile
bed versus gas velocity with liquid velocity
as parameter 80
26 Experimental pressure drop of a 3-stage
mobile bed with hardware screen support 83
27 The variation of pressure drop with liquid-
to-gas ratio and air velocity 85
28 Effects of bed depth and number of stages on
pressure drop 86
29 Effect of packing diameter on pressure drop 87
30 Pressure drop vs. superficial gas velocity
for 3-stage mobile bed with plastic net
support 88
31 Pressure drop across four plastic net
supports 90
32 Pressure drop due to liquid holdup and
weight of packing 91
VII
-------
FIGURES (continued)
Number Page
33 Experimental grade penetration curves 92
34 Experimental grade penetration curves 93
35 Experimental grade penetration curves 94
36 1-stage and 3-stage mobile bed scrubber
performance 96
37 Penetration curves from interstage
sampling 97
38 Experimental cut/power relationship
for the mobile bed scrubber 99
39 Comparison between published mobile bed
performance data with present study 100
40 The variation of bed expansion with
liquid-to-gas ratio 102
41 The variation of bed expansion with
liquid-to-gas ratio 103
42 Effect of packing diameter on bed
expansion 104
43 The variation of minimum fluidization
velocity with liquid flow rate 106
44 The variation of minimum fluidization
velocity with packing sphere diameter 107
45 Effects of slurry on pressure drop 108
46 Slurry scrubbing test data 109
47 F/C scrubbing pressure drop Ill
48 The variation of cut diameter with
condensation ratio 112
49 Penetration for 1.0 umA diameter
particle versus condensation ratio 113
Vlll
-------
FIGURES (continued)
Number Page
50 The grown particle experimental setup 118
51 The predicted and measured particle
grown size distribution 121
52 Typical pressure drop - flow characteristics
in conventional packed towers 124
53 Region of mobile bed operation mode 126
54 Predicted and measured pressure drop
(Chen and Douglas's correlation) 143
55 Predicted and measured pressure drop
(Wofniak's correlation) 144
56 Predicted and measured pressure drop
(Kito et al. ' s correlation) 145
57 Predicted and measured pressure drop
(Uchida et al.'s correlation) 146
58 Measured and predicted pressure drop for
a three stage mobile bed scrubber with
hardware screen support, and each stage
packed with 3.8 cm dia. spheres to a
depth of 23 cm 147
59 Comparison between measured and predicted
mobile bed pressure drop 150
60 Pressure drop through a single stage of a
mobile bed obtained at EPA/TVA Shawnee
plant 151
61 Generalized F/C scrubber system 154
62 Scrubber penetration for different
scrubber stages 166
63 Initial and grown particle size
distribution 168
64 The penetration curve for Runs No.
FC-7 and FC-8 169
IX
-------
FIGURES (continued)
Number Page
65 The penetration curve for Runs
No. FC-7 and FC-8 .................. 170
66 Predicted and measured particle
penetration for Run No. FC-23 ............ 171
67 Predicted and measured penetration for 1.0 ymA
diameter particle versus condensation ratio ..... 172
68 Predicted and measured variation of cut
diameter with condensation ratio .......... 173
69 Cut/power relationship for scrubbers ........ 175
70 Comparison between mobile F/C and sieve
plate F/C scrubber performance ........... 178
71 Comparison between mobile bed F/C and sieve
plate F/C scrubber performance ........... 179
72 Comparison between mobile bed F/C and spray
F/C scrubber performance ..............
73 Typical process design of a F/C scrubber
system ....................... 182
74 Typical fly ash distribution and grown size
distribution .................... 186
75 Predicted mobile bed and F/C mobile bed
scrubber performance ................ 187
Appendices
C-l through C-67 Experimental grade penetration curves. . 218-234
D-l through D-4 Experimental penetration curves ..... 236
-------
FIGURES (continued)
Number Page
E-l The penetration curve for Runs No. FC-1,
FC-2, and FC-3 238
E-2 The penetration curve for Runs No. FC-4
and FC-5 238
E-3 The penetration curve for Run No. FC-6 238
E-4 The penetration curve for Runs No. FC-7
and FC-8 238
E-5 The penetration curve for Runs No. FC-9
and FC-10 239
E-6 The penetration curve for Runs No. FC-11,
FC-12, and FC-13 239
E-7 The penetration curve for Runs No. FC-14,
FC-15, and FC-16 239
E-8 The penetration curve for Runs No. FC-17
and FC-20 239
E-9 The penetration curve for Runs No. FC-18
and FC-19 240
E-10 The penetration curve for Runs No. FC-21
and FC-22 240
E-ll The penetration curves for Runs No. FC-23,
FC-24 and FC-25 240
E-12 The penetration curves for Runs No. FC-26
and FC-27 240
XI
-------
TABLES
Number Page
1 Pressure Drop Correlations 8
2 Cut/Power Relationships 13
3 Drop Diameter Interval for DC-1 Drop Counter .... 25
4 Preliminary Entrainment Measurement Scrubber
Operating Conditions 31
5 DC-1 Drop Counter Entrainment Data and Scrubber
Operating Conditions 46
6 List of Pilot Plant Components 57
7 Flowrates and Conditions for Mobile Bed F/C
Scrubber Pilot Plant 58
8 Particle Count Results 115
9 Particle Growth Data 120
Appendices
A-l DC-1 Drop Counter Data for Run No. DC-1 203
A-2 DC-1 Drop Counter Data for Run No. DC-2 203
A-3 DC-1 Drop Counter Data for Run No. DC-3 203
A-4 DC-1 Drop Counter Data for Run No. DC-4 203
A-5 DC-1 Drop Counter Data for Run No. DC-5 204
A-6 DC-1 Drop Counter Data for Run No. DC-6 204
A-7 DC-1 Drop Counter Data for Run No. DC-7 204
A-8 DC-1 Drop Counter Data for Run No. DC-8 204
A-9 DC-1 Drop Counter Data for Run No. DC-9 205
XII
-------
TABLES (continued)
Number Page
A-10 DC-1 Drop Counter Data for Run No. FC-10 205
A-ll DC-1 Drop Counter Data for Run No. FC-11 205
A-12 DC-1 Drop Counter Data for Run No. FC-12 205
A-13 DC-1 Drop Counter Data for Run No. FC-13 206
A-14 DC-1 Drop Counter Data for Run No. FC-14 206
A-15 DC-1 Drop Counter Data for Run No. FC-15 206
A-16 DC-1 Drop Counter Data for Run No. FC-16 206
A-17 DC-1 Drop Counter Data for Run No. FC-17 207
A-18 DC-1 Drop Counter Data for Run No. FC-18 207
A-19 DC-1 Drop Counter Data for Run No. FC-19 207
A-20 DC-1 Drop Counter Data for Run No. FC-20 207
A-21 DC-1 Drop Counter Data for Run No. FC-21 208
A-22 DC-1 Drop Counter Data for Run No. FC-22 208
A-23 DC-1 Drop Counter Data for Run No. FC-23 208
A-24 DC-1 Drop Counter Data for Run No. FC-24 208
A-25 DC-1 Drop Counter Data for Run No. FC-25 209
A-26 DC-1 Drop Counter Data for Run No. FC-26 209
A-27 DC-1 Drop Counter Data for Run No. FC-27 209
A-28 DC-1 Drop Counter Data for Run No. FC-28 209
B-l Test Conditions and Particle Data 211
B-2 Test Conditions and Particle Data 211
B-3 Test Conditions and Particle Data 211
Xlll
-------
TABLES (continued)
Number Page
B-4 Test Conditions and Particle Data 212
B-5 Test Conditions and Particle Data 213
B-6 Test Conditions and Particle Data 213
B-7 Test Conditions and Particle Data 214
B-8 Test Conditions and Particle Data 214
B-9 Slurry Scrubbing Test Conditions and
Particle Data 215
B-10 F/C Scrubbing Test Conditions and
Particle Data 216
B-ll F/C Scrubbing Test Conditions and
Particle Data 216
xiv
-------
ABBREVIATIONS AND SYMBOLS
A = cross-sectional area of scrubber, cm2
a = interfacial area of packing, cm2/cm3
a^ = geometrical surface area of static bed per unit volume
of static packing, cm2/cm3
at = interfacial area for transfer volume of scrubber, cm2/cm3
C' = Cunningham slip factor, dimensionless
c = salt concentration in solution, g/cm3
c = particle mass loading, g/DNm3
c . = inlet particle mass loading, g/DNm3
c = outlet particle mass loading, g/DNm3
c = heat capacity of particle, cal/g-°K
c = total particle loading, g/DNm3
pt
D = column diameter, cm
Dp = Diffusivity of water vapor in carrier gas, cm2/s
d, = ball or packing diameter, cm
d, = drop diameter, vim or cm
d,M = mass median drop diameter, ym or cm
djN = number median drop diameter, ym or cm
d = equivalent diameter of the grid opening, cm
\2
d, = hole diameter, cm
d = mass mean diameter, cm or ym
d = aerodynamic particle diameter, ymA
d = grown aerodynamic particle diameter, ymA
pa-2
d = original physical particle diameter, ym
d = grown physical particle diameter, ym
P2
d = cut diameter, ymA
pc ' M
d' = physical geometric mass median diameter, cm
Jr o
d = hot wire diameter, cm
w '
F = foam density, dimensionless
f = fraction of water vapor condensing on particles, fraction
xv
-------
fs ~ fractional °Pen area of the supporting grid, fraction
fy = mole ratio of water vapor condensed, fraction
G = gas mass velocity, g/cm2-s
Gm£ = minimum fluidization mass flow, g/hr-cm2
g = acceleration of gravity, cm/s2
H = packed column height or the distance between retaining
grids, cm
Hi = dynamic bed height, cm
H, = height of liquid column retained on the supporting grid, cm
H = static bed height, cm
H! = humidity in the saturated inlet gas, g/g
h, = packing holdup, cm3/cm3
h^ = gas holdup, cm3/cm3
h/-, = heat transfer coefficient, kcal/cm3-s-°C
tj
hT = liquid holdup in bed, cm3/cm3
Li
h, = liquid holdup based on fixed bed, cm3/cm3
h r = particle to gas heat transfer coefficient, cal/cm2-s-°K
h = packing holdup related to fixed bed, cm3/cm3
K = inertial impaction parameter, dimensionless
k = constant characterizing the fraction of liquid being
atomized, --
k = thermal conductivity of gas, cal/cm2-s-°K/cm
k,, = mass transfer coefficient, g/cm3-s-atm
k'~ = particle to gas mass transfer coefficient, gmol/cm2-s-atm
ki = mass transfer coefficient, gas to liquid, gmol/cm2-s-atm
L = liquid mass velocity, g/cm2-hr-atm
LM = latent heat of vaporization for water, cal/gmol
Mj = molecular weight of water, g/mol
M2 = molecular weight of nontransferring gas, g/mol
m = total mass of particles, g
N = total number of particles, #
N-
= number of drops counted in the i'th bin, number
n = number of mobile bed stages, --
ni = drop concentration corresponding to i'th bin, #/cm
n = particle number concentration, #/cm3
xvi
-------
Pt = overall penetration, fraction or percent
Pt = penetration due to diffusiophoresis, fraction
Ft, = particle penetration for particle diameter, d , fraction
Q. p 3.
PBM = mean partial pressure to nontransferring gas, atm
Pp = water vapor partial pressure in bulk of gas bubble, atm
p . = water vapor partial pressure at vapor liquid, interface, atm
q' = condensation ratio, g/g
r = distance in the direction of diffusion, cm
r = particle radius, cm
Tp = gas bulk temperature, °K
T^ = temperature of liquid bulk, °K
T . = particle interface temperature, °K
t = time, s
u- = superficial gas velocity, cm/s
Up = original entrainment velocity in wetted packing, cm/s
Up£ = flooding velocity, cm/s
u-. = gas velocity in bed, cm/s
Up r = minimum fluidization velocity, cm/s
ur ^1 = minimum fluidization velocity of dry packing, cm/s
UT = superficial liquid velocity, cm/s
Li
upn = particle deposition by diffusiophoresis, cm/s
V = volume of gas sampled, cm
V - = gas volume sampled by the i'th bin, cm3
y = mole fraction water vapor, fraction
Z = coordinate axis, cm
Latin
AP = pressure drop, cm W.C.
AP^ = pressure drop due to the weight of dry packing, cm W.C.
AP = column friction loss, cm W.C.
L^
APf = supporting grid friction loss, cm W.C.
APL = pressure drop due to liquid froth retained on the
supporting grid, cm W.C.
APLh = Pressure dr°P due to liquid holdup in bed, cm W.C.
AP
w
= overall pressure drop across the scrubber, cm W.C.
xv 11
-------
p, = packing density, g/cm3
PG = gas density, g/cm3
p, = liquid density, g/cm3
PM = molar density of water, gmol/cm3
p = particle density, g/cm3
p = original particle density, g/cm3
p = density of grown particle, g/cm3
Pw = density of water, g/cm3
a = geometric standard deviation, dimensionless
o
a1 = physical size geometric standard deviation,dimensionless
o
a. = surface tension of liquid, dyne/cm
e = porosity of static bed, fraction
e £ = voidage of bed at minimum fluidization velocity,
dimensionless
UQ = gas viscosity, g/cm2-s
y^ = liquid viscosity, g/cm2-s
E,f = experimental coefficient, dimensionless
xvi 11
-------
ACKNOWLEDGEMENT
Air Pollution Technology, Inc. wishes to express its ap-
preciation to Mr. Dale Harmon, E.P.A. Project Officer; Dr.
Leslie E. Sparks, E.P.A. previous Project Officer; and Mr.
James Abbott, E.P.A. for excellent technical coordination and
very helpful assistance in support of our technical effort.
xix
-------
SECTION 1
SUMMARY AND CONCLUSIONS
SUMMARY
A mobile bed scrubber is a three-phase contacting device
which consists of several packed beds stacked inside a vessel
shell. Gas flowing upward with a high velocity fluidizes
the packing and liquid simultaneously flows downward through
the column. The packing commonly used is 2.5 cm or 3.8 cm
diameter lightweight spheres but other materials, such as
lightweight plastic rings have been used (Levesh, 1968) . The
packing is supported and retained by nonflooding grids suffi-
ciently far enough apart to permit turbulent and random motion of
the spheres . The static packing depth is 15-60 cm (0.5 - 2 ft).
Mobile bed contactors have been used as gas absorption
towers for many years. In recent years, they are also being
used in the removal of sulfur dioxide and particulates from
stack gases by a lime or limestone slurry. Fundamental studies
on the mechanics of mobile beds such as hydrodynamics, pressure
drop, liquid holdup, minimum fluidization velocity and axial
liquid mixing have been conducted by a number of investigators.
Prior to this work there had been no carefully performed study
on the particle collection in a mobile bed scrubber.
The program which is reported here was undertaken in order
to develop better information on the characteristics of the
mobile bed scrubber performance and to point the way to the
improvement of mobile bed scrubber design. Both experimental
and theoretical investigations of mobile bed scrubber perform-
ance were made.
The objectives of this study were to:
1. Conduct an experimental study aimed at obtaining
-------
liquid entrainment data for the proper design of the
entrainment separator.
2. Conduct an experimental study on particulate scrubbing
in mobile bed scrubbers
3. Determine the effects of lime or limestone slurry
on mobile bed performance.
4. Conduct an experimental study of flux force/condensa-
tion (F/C) scrubbing in mobile bed scrubbers.
5. Develop engineering design equations.
6. Evaluate mobile bed scrubbers.
7. Develop specific research and development recommen-
dations .
Experimental Study
Two pilot plants were constructed. One was specifically
for the study of liquid entrainment from mobile bed scrubbers.
It had a maximum gas flow capacity of 56.6 m3/min (2,000 CFM)
and consisted of a single stage mobile bed scrubber, a blower,
pump, supply and catch tank, and auxiliary equipment. The
experiments were done with air and water at ambient temperatures.
Measurements included entrainment loading and drop size dis-
tribution for various gas and liquid flow rate combinations.
A hot wire anemometer and cascade impactor were used for
drop sampling.
The second pilot plant was composed principally of a
furnace, spray-type quencher, mobile bed scrubber, spray-type
cooling tower, and two induced draft fans. The mobile bed
scrubber could have a maximum of three stages housed in the
scrubber shell and was designed for a maximum gas velocity of
51 m3/min (1,800 CFM).
The pilot plant was operated in three modes: cold operation,
slurry scrubbing, and F/C scrubbing. For cold operation the
burner, furnace, quencher, and the cooling tower were idle.
The air and water were at ambient conditions and no precondi-
tioning was applied to either one. The effects on particle
-------
collection efficiency of such factors as number of mobile bed
stage, bed height, gas velocity, liquid-to-gas ratio, packing
size, supporting grid types, and types of aerosol were deter-
mined experimentally. Particle measurements included the size
distribution and concentration. Particle collection efficiency
was calculated from particle data and was reported in terms of
grade penetration curve.
Some experiments were made with limestone slurry in place
of clear water and the effect of this change on particle col-
lection efficiency was determined. Particle size distribution
and concentration at the scrubber inlet and outlet were mea-
sured with cascade impactors. The particle penetration was
computed and was related to:
1. Failure of the scrubber to collect particles.
2. Failure of the entrainment separator to collect
entrained drops.
3. Particle generation by droplet evaporation.
Water condensation effects enhance fine particle scrubber
collection efficiency and their influence on the mobile bed
scrubber was studied. F/C scrubbing involves the saturation
of the gas with water vapor and the subsequent inducement of
condensation of water vapor on particles before the major
collection mechanism occurs. Because condensation of water
vapor on the particles increases their mass, they become more
susceptible to collection by inertial impaction.
F/C scrubbing is applicable in situations where the gas
is hot or where low cost waste steam is available. In the
present study, hot gas was obtained by heating the air in the
furnace with the natural gas burner. The gas was then saturated
with water vapor in the spray quencher. Condensation of water
vapor on particles was induced in the mobile bed scrubber by
contacting the hot and humid gas with cold water. The improve-
ment in collection efficiency depends on the amount of vapor
condensed and the particle number concentration. These two
parameters, along with the usual particle size distribution and
-------
concentration measurements, were determined for all experiments
Experimental Findings
Liquid Entrainment -
The liquid entrainment loading measured with the hot wire
anemometer was much lower than that measured with cascade
impactors. Both devices measured the entrainment flow rate
which increased with both the gas flow and liquid flow rates.
At a constant superficial gas velocity the entrainment flow
rate increased gradually with increasing liquid/gas ratio
up to the flooding condition of the mobile bed. Further
increases in the liquid/gas ratio beyond this value sharply
increased the entrainment flow rate.
For a mobile bed scrubber operating under typical indus-
trial conditions (u ' = 2.7 m/s and Qj/QG = 6-7 ^M3), the hot
wire anemometer measured an entrainment flow rate of about
0.1 £/min-m2. The measured drop size distribution did not
change much with operating conditions. The drop size distribu-
tion measured with the hot wire anemometer followed the log-
normal distribution. The mass median diameter was about 250
ym and the geometric standard deviation was 1.6.
Particle Collection -
Particle collection data for all experimental runs were
reduced to plots of particle penetration versus particle dia-
meter, commonly called grade penetration curve. Particle
penetration depended only on the overall pressure drop across
the scrubber and was independent of the number of scrubber
stages. The grade penetration curve of a 1-stage mobile bed
was the same as that of a 2-stage mobile bed if the overall
pressure drops were the same.
-------
Aerosol types had no significant effect on the scrubber col
lection efficiency.
A cut/power relationship has been established in this study
for mobile bed scrubbers. It can be described by the following
equation:
dpc = 60.3 ( APw)- (1)
where d = scrubber performance cut diameter, ymA
AP = overall pressure drop across the scrubber, cm W.C.
The performance data obtained in this study are consistent
with results obtained by other researchers.
Slurry Scrubbing -
The use of slurry as the scrubbing liquid did not affect
the scrubber efficiency. A cut/power relationship for the
mobile bed using slurry was identical to that determined in the
cold operation mode. This indicates that increasing the liquid
density by addition of limestone to the scrubber liquid does
not influence the dependency of particle collection efficiency
on pressure drop.
In some industrial scrubbers, the use of slurry is reported
to result in lower efficiencies. This decrease in efficiency
could be due to the failure of the entrainment separator to
collect entrained slurry drops. In the present study, the tube
bank entrainment separator had high efficiency and entrained
solids were negligible.
F/C Scrubbing -
The use of F/C conditions greatly improved the collection
efficiency for submicron particles. At a scrubber pressure drop
of 13 cm W.C., the collection efficiency of the mobile bed
scrubber with no F/C effect for 1 ymA diameter was less than 5%.
-------
At the same pressure drop, the collection efficiency increased
to 50% with a condensation ratio of 0.2 g/g.
Besides pressure drop, the collection efficiency of the
mobile bed F/C scrubber also depended on condensation ratio
and particle number concentration. High condensation ratio
coupled with low particle number concentration resulted in
larger grown particle size, which were more susceptible to
collection by inertial impaction.
Pressure Drop -
The pressure drop characteristics, which represent most
of the power requirement of the mobile bed, were determined
experimentally in this study. The major findings are:
1. Retaining grid geometry has a profound effect on
scrubber pressure drop. Under the same operating
conditions,' support with small openings and
fractional open area caused a higher pressure
drop. The higher pressure drop is probably
due to the liquid froth retained on the grid.
2. Pressure drop increases with increasing static
bed height, density of packing material, super-
ficial gas velocity, and superficial liquid velo-
city; but it decreases with increasing packing size.
For a fully fluidized mobile bed operating below
flooding conditions, the dependence of pressure
drop on gas velocity is small.
3. The measured pressure drop across the mo.bile bed
with large opening supports agreed with those reported
by Douglas and Snider (1963) and by Pollock et al.
(1967).
Engineering Design Equations
Pressure Drop -
For a fully fluidized mobile bed, the pressure drop is made
up of the sum of those due to the dry retaining grids, the weight
of the dry packings, the liquid holdup in the bed, liquid froth
-------
retained on the grid, and the wall friction. Except for the
grids with small openings and small fractional open area, the
losses due to wall friction and supporting grid friction are
small. There is also little liquid retained on the supporting
grids. The pressure drop across the mobile bed then can be
considered equal to the sum of the weight of the packing and
the liquid holdup in the bed.
There are several empirical correlations reported in the
literature for the prediction of pressure drop across a mobile
bed. Table 1 lists four of the correlations for the calcula-
tion of pressure drop due to the weights of dry packing and
liquid holdup. There are other correlations, mostly by Russian
researchers, which contained constants which have to be deter-
mined experimentally for each application.
Predictions by these correlations were compared with the
experimental data obtained in this study. The comparison
revealed that predictions by Uchida, et al.'s correlation are
much higher than that actually measured for the scrubber with
large opening support grids. Predictions by Kito, et al.'s
correlation and by Wozniak's correlation are better than
Uchida, et al.'s correlation, but they still are higher than
that measured.
Chen and Douglas' prediction agrees with the data for the
scrubber with large opening support grids. However, Chen and
Douglas' correlation did not predict the correct dependence
of pressure drop on liquid flow rate. Compared to the experi-
mental pressure drop data for the mobile bed with small opening
supporting grids, Chen and Douglas' predictions are much
lower.
Since predictions by the pressure drop correlations reported
in the literature do not agree with the data obtained in this
study, an empirical equation was developed through dimensional
analysis. The agreement between predictions by this equation
with data reported in the literature is good. The empirical
equation is:
-------
TABLE 1. PRESSURE DROP CORRELATIONS
Investigators
Chen § Douglas
(1969)
Wozhiak
(1977)
Kito, et al .
(1976)
Uchida, et al .
(1977)
Correlation
APW = (1-0 Pb Hs + hLo PL Hs
h = 0.0944 d ~°-5 L°'5 +0.02
Lo D
APW = (1-e) Pb Hs + APLh
(TT \ 0.4515 /J ,, _ \1,798
n \ 1 CU Ur pr \
si | b G G |
"11 1
V \ »G /
/dbuL^L\"26'
( ^ )
APw = Cl-0 Pb Hs + hLo pL Hs
/f d \-°-^
h -0 06 + 0 03 16 I I ^-0=84 n 0,18 TT- QM
Lo 1 D, I b Kb s L
\ c/
APw = ^-^ Pb Hs + APLh
/d \-°-8!t
AP = 3317 y 2'3 £ -»J'2/_e\ d -0.8, 0.18 H
Ln L s Iplb^bsL
\ c/
-------
APW = n (APb + APLh + APL) + (n + 1) APf + AP (2)
APb = (1-0 Pb Hs C (3)
„ ,, . - 1.75 I A \ 0-96
APT, = 83.0 pru 2
Lh G b \cL i \ y
u G i, i, ,4^
/ D \2-6
APT = 9.2 xlO'9 (- c, u.05 UT2'° (5)
L \±s e' G L
where: AP = overall scrubber pressure drop, cm W.C.
AP^ = pressure drop due to weight of dry packing, cm W.C.
APriL = pressure drop due to liquid holdup in bed, cm W.C.
AP, = pressure drop due to liquid froth retained on
grid, cm W.C.
APf = grid friction loss, cm W.C.
AP = column friction loss, cm W.C.
D = column diameter, cm
c '
d, = diameter of the packing, cm
d = equivalent diameter of the grid opening, cm
f = fractional open area of the retaining grid,
dimens ionless
H = static bed height, cm
o
n = number of mobile bed stages,
Up = superficial gas velocity, cm/s
UT = superficial liquid velocity, cm/s
LI
p, = density of packing material, g/cm3
pp = gas density, g/cm3
PT = liquid density, g/cm3
e = porosity of static bed, fraction
yp = gas viscosity, g/cm -s
yT = liquid viscosity, g/cm2-s
"AP,:" and "AP " can be calculated by methods presented by
Perry (1973).
-------
Particle Collection -
There is only one model available in the literature for
particle collection in a mobile bed. It is a semi-empirical
relationship presented by Bechtel Corporation in a June 1971
report on the Shawnee project for EPA and cited by Calvert,
et al. (1972). This equation was based on the premise that
particle collection is due to inertial impaction on balls.
Predictions of efficiency by this equation did not agree
with performance data and were lower than measured. The
Bechtel equation is not based on a realistic model of the
particle collection mechanisms and does not account for the
influence of the support grid. As mentioned earlier, the
scrubber collection 'efficiency depends only on the pressure
drop. The mobile bed scrubber with small opening supporting
grids has higher pressure drop than the mobile bed with large
opening supports. Consequently, the collection efficiency is
higher for the mobile bed with small opening supports.
Particle collection in a mobile bed scrubber is partly
due to inertial impaction on the atomized liquid drops. Thus,
particle collection by this mechanism can be predicted if the
atomized drop diameter and amount of liquid in drop form are
known. However, the hydrodynamics of a three phase fluidized
bed are extremely complex, and it was not possible to derive
theoretical equations for predicting drop size and quantity
of drops. Empirical approximations based on pressure drop
relationships for gas atomized scrubbers were used to predict
collection efficiency without success.
A new empirical equation was developed for predicting
particle collection in a mobile bed scrubber. The equation
developed in the present study is:
Ptd = exp
- 9.84 x 10 -" TAP ) l 96 d 1-6
w Pa
(6)
10
-------
where: d = aerodynamic particle diameter, ymA
pa
AP = overall pressure drop, cm W.C.
Pt^ = particle penetration for particle diameter
"d ", fraction
pa
In order to predict the particle collection efficiency
of a mobile bed, one first predicts the pressure drop by
means of equation (2) and then the performance cut diameter
by means of equation (1) or (6) . The penetration for other
particle diameters can be calculated from equation (6) .
In calculating the collection of one stage of a multi-
stage scrubber, one first calculates "Ft," for the multi-
stage scrubber, and then calculates one stage penetration
according to the following equation:
)n = Pt
d
where Ptj = penetration for one stage of a multi-stage scrubber,
fraction
n = number of stages, -
F/C Scrubbing-
The general theory on F/C scrubbing has been developed
by Calvert, et al. (1973) and has been modified and simplified
by Calvert and Gandhi (1977) . By combing the theory for F/C
scrubbing and the design equation for the mobile bed, the
following equations were obtained which can be used to predict
the particle collection performance of a F/C mobile bed scrubber.
Case 1: Condensation and particle growth occurred before the
mobile bed scrubber.
,
exp(-9. 84x10-" AP d ) (3)
P 2
!,6 -i
)
J
0.5
"pa2 ~ +
11
-------
fp q' + 3.78 x 10
- 1 0
P2 £
' + 3.78X 10'10 np dpi3 ppi
(10a)
0.165 + (0.0272 + 4 d
„
Pd P
pi
d =
pl
722 ^ np PP2J
(10b)
x 10* (lOc)
where
pa2
V
V
d '
P
V
H =
Ptd =
q' =
AP =
w
Pi
grown aerodynamic particle diameter, ymA
grown physical particle diameter,ym
original aerodynamic particle diameter, ymA
original physical particle diameter, ym
fraction of water vapor condensing on particles,
fraction
humidity in the saturated inlet gas, g/g
particle number concentration, #/cm3
particle penetration for diameter, d , fraction
pa
condensation ratio, g/g
overall pressure drop across mobile bed scrubber,
cm W.C.
initial particle density, g/cm3
density of grown particle, g/cm3
Case 2: Condensation and particle growth occurred within the
mobile bed scrubber.
Ptd =
1 -
0.85q' (1 - f )
0.62 + H,
expI-9 .84x10"
exp(-9.84 x
1>96 d l
w pa
AP 1-96 d
w pa2
n-1
n
n
(ID
12
-------
where "d " is given by equation (9)
pa 2
n = number of mobile bed stages, -
For a "£ " of 0.15, which was determined experimentally,
the predicted collection efficiency is slightly lower than
that measured for large particles.
Evaluation of Mobile Bed Scrubber
Each scrubber type has a unique cut/power relationship
which describes the dependency of scrubber performance on
pressure drop. The cut/power relationships for other scrubber
types are listed in Table 2 and plotted in Figure 1. The
uniqueness of the cut/power relationship offers a simple
method to evaluate the capabilities of different scrubber
types.
TABLE 2. CUT/POWER RELATIONSHIPS
Gas-Atomized Scrubber: d = 3.53 (AP)-°'5
Packed Bed with 2.5 cm Rings: d = 5.8 (AP)"0"2
Sieve Plate with F = 0.4 and dh = 0.32 cm: d =2.56 (AP)"0'255
Note: d = performance cut diameter,
AP = pressure drop, cm W.C.
13
-------
10
i i r
of,
w
2
i—i
o
H
U
- MOBILE BED
3a -
PACKED BED WITH 2.5 cm
DIA. RINGS OR SADDLES
SIEVE PLATE WITH FOAM
DENSITY OF 0.4 g/cm3
AND 0.51 cm HOLE
DIAMETER
- SAME AS 3a EXCEPT
0.32 cm HOLE DIAMETER
- GAS-ATOMIZED SPRAY
J L
10 30
PRESSURE DROP, cm W.C.
Figure 1. Cut/power relationship for scrubbers.
100
14
-------
Of the four scrubber types (gas-atomized spray, mobile
bed, packed bed, and sieve plates), the gas-atomized spray
scrubber has the best performance capability because it can
attain the smallest performance cut diameter with the same
pressure drop. At a pressure drop of 20 cm W.C., the gas-
atomized spray scrubber can achieve a performance cut dia-
meter of 0.75 ymA. At the same pressure drop, the performance
cut diameters are 1. 6 ymA, 1. 7 ymA, and 1. 2 ymA for the mobile bed,
packed bed, and sieve plate with 0.32 cm hole diameter, re-
spectively. Industrial mobile beds usually operate at a
pressure drop around 30 cm W.C. The performance cut diameter
is approximately 0.9 ymA which is better than that of packed
bed and sieve plate; but it is still larger than that of the
venturi scrubber.
In F/C scrubbing, particle growth may be induced before
or within the scrubber. A condenser is required if particle
growth occurred before the scrubber. In this case, any type
of particle scrubber may be used after the condenser to collect
the grown particles. Since the gas-atomized scrubber has the
best performance characteristics, it is the logical choice
for this arrangement.
In the situation where condensation and particle growth
are induced within the scrubber, the scrubber should have high
heat and mass transfer capability. Suitable scrubbers include
the mobile bed, sieve plate, packed bed, and spray.
The performance of the F/C sieve plate and F/C spray has
been determined by Calvert,et al. (1975) through pilot plant
studies. When comparing their results with the F/C mobile bed
scrubber performace determined in this study, it shows that the
F/C mobile bed has the best characteristics among the three.
The capital outlay for the three F/C scrubber systems is
approximately the same. However, the mobile bed has the highest
annual operating power cost because it has a larger cooling
tower.
15
-------
CONCLUSIONS
The principal objectives of this study were achieved. The
following conclusions can be drawn, based on evaluation of experi
mental results.
Particle Collection
1. The particle collection capability of the mobile bed
scrubber is a strong function of the gas phase pressure
drop. The higher the pressure drop, the higher the
collection efficiency will be. Any parameter that in-
creases the scrubber pressure drop (except friction
losses) will increase the scrubber collection efficiency,
The cut/power relationship is given in equation (2).
2. Types of aerosol have no significant effect on the
particle collection mechanisms and efficiency.
3. The use of limestone slurry does not affect the
collection efficiency.
4. The performance capability of the mobile bed scrubber
is less than that of the gas-atomized spray scrubber
with the same pressure drop. Compared to the packed
bed and sieve plate, the mobile bed has better per-
formance capability when the pressure drop is above
20 cm W.C.
F/C Scrubbing
1. The use of F/C effect improve the collection efficiency
for submicron particles. The extent of improvement
depends on the condensation ratio and particle number
concentration. High condensation ratio combined with
low particle number concentration leads to high
collection efficiency.
2. The fraction of condensing vapor condensed on the
particles was 0.15 for particle number concentrations
of 1 x 107 to 5 x 107/cm3.
16
-------
Pressure Drop
The pressure drop across a fully fluidized mobile bed is
made up of the sum of those due to the dry retaining grids,
the weight of the dry packings, the liquid holdup in the bed,
the liquid froth retained on the grid, and the wall friction.
It increases with increasing static bed height, density of the
packing material, superficial gas velocity, and superficial
liquid velocity, but it decreases with increasing packing size.
Retaining grid geometry has a significant effect on the pressure
drop. The use of grids with small openings and small fractional
area increases the liquid froth retention on the grid which in
turn increases the pressure drop.
Engineering Design Equations
The correlations reported in the literature for scrubber
pressure drop and particle collection did not agree with data
obtained in this study. New equations were derived for
predicting scrubber pressure drop and particle collection.
These equations compare favorably with available data re-
ported in the literature, as well as data from this study.
Potential as F/C Scrubber
1. As a F/C scrubber, the mobile bed scrubber can have
better collection efficiency than sieve plate and
spray scrubbers. The capital cost of a mobile bed
F/C scrubber system is slightly higher than that
for the spray F/C scrubber system, but it is slightly
lower than that for the sieve plate system. The
mobile bed F/C scrubber system has the highest
annual power cost among the three systems. The
overall annualized cost of the three F/C systems
is about the same.
17
-------
The condensation of water vapor on particles may
be induced to occur before or within the scrubber.
In the first arrangement, the venturi scrubber is
a better choice as the particle scrubber. In the
second arrangement, the mobile bed appears to be
better.
Due to the low attainable condensation ratio, the
mobile bed scrubber may be used as a SO and particle
A.
scrubber in power plant applications, but is not
preferred as a F/C scrubber.
18
-------
SECTION 2
INTRODUCTION
A mobile bed scrubber is a counter-current contacting tower
in which beds of low density spheres are fluidized by the upward
flowing gas and are irrigated by the downward flowing liquid.
The spheres may typically be hollow plastic spheres of from 1 to
4 cm diameter and of density considerably below that of water.
The beds are supported and retained by nonflooding grids.
Such a device was first described in the literature by
Kielback in 1959 and later a modified version was developed by
Douglas et al. (1963). The primary difference between the two
versions of the mobile bed scrubber is that in the Kielback
design, the distance between the lower and upper retaining grid
is only slightly larger than the static bed height while in the
configuration of Douglas et al. , the static bed height is much
less than the distance between grids and the bed is greatly
expanded during operation.
In industry, Kielback's design is termed "floating bed
scrubber" and the design by Douglas et al. is termed "turbulent
contact absorber." Other names like "mobile bed scrubber,"
"fluidized packing contactor," and "fluidized bed scrubber" have
been used to describe a packed tower operating in a fluidized
state. In this report, the term "mobile bed scrubber" is adopted.
Mobile bed scrubbers have been used for scrubbing flue gas
from coal-fired utility boilers. Experience with this application
has pointed to the problem of liquid entrainment. Entrainment
from mobile bed scrubbers has caused the plugging of entrainment
separator and ductwork, failure of fan, and excessive emission of
particulates.
19
-------
Nearly 100% removal of entrained drops is essential for
trouble-free operation of the scrubber system. Design of high
efficiency entrainment separators requires knowledge of the
entrained drop size distribution and concentration. Calvert et al.
(1977) have started the measurement of entrainment from mobile
bed scrubbers and have provided some information on drop size
distribution and concentration. Their work is continued in this
study.
Fundamental studies on the mechanics of mobile beds such as
hydrodynamics, pressure drop, liquid holdup, minimum fluidization
velocity, and axial liquid mixing have been conducted by a number
of investigators. There is no carefully performed study on the
particle collection in a mobile bed scrubber. Particle collection
efficiency data reported in the literature varied as much as two
orders of magnitude under similar operating conditions. A systematic
pilot study was carried out to investigate the particle collection
in a mobile bed scrubber.
Present day scrubbers are hampered by the large energy
requirement for high removal efficiencies on particles in the size
range of 0.1 to 2 microns in diameter. This is due to the decreased
effectiveness of the inertial and diffusional collection mechanisms
for particles in this size range. Flux force/condensation (F/C)
effects resulting from water vapor condensation can greatly improve
the collection efficiencies of low energy scrubbers for fine
particles.
In this report, flux forces are defined as those caused by
thermophoresis and diffusiophoresis, but not electrophoresis .
Accordingly, we consider only those F/C scrubbers where particle
removal from the gas is aided by temperature or vapor concentra-
tion gradients and particle growth is due to vapor condensation.
These effects can result from the cooling of a hot, humid gas by
contact with cold liquid, the condensation of injected steam, or
other means.
The F/C sieve plate scrubber and F/C spray scrubber have been
studied in detail by Calvert et al. (1975). The feasibility
20
-------
of operating the mobile bed scrubber as a F/C scrubber is inves-
tigated in the present study.
This report presents the results of the experimental in-
vestigation of entrainment, particle collection, slurry scrub-
bing, and F/C scrubbing of mobile bed scrubbers. Mobile bed
scrubber design recommendations and operating practices to
ensure adequate fine particle collection are established from
data analysis.
The economics and performance of F/C scrubbing using a
mobile bed scrubber are evaluated and compared with the F/C spray
scrubber and F/C sieve plate scrubber. The potential application
of F/C mobile bed scrubbers in coal-fired power plants is
evaluated.
21
-------
SECTION 3
PRELIMINARY STUDIES
Design of high efficiency entrainment separators requires
knowledge of the entrained drop size distribution and concen-
tration. Data on size distribution for drops smaller than 20 ym
diameter are especially important, for without it the design of
high efficiency entrainment separators is impossible. A prelim-
inary study was undertaken to provide data needed to design an
efficient entrainment separator for the mobile bed F/C pilot
plant.
ENTRAINMENT MEASUREMENTS
Equipment
A pilot plant was constructed to study the liquid entrain-
ment. It consisted of a blower, a mobile bed scrubber, a knitted
mesh entrainment separator, a scrubber sump tank, and a pump.
Figure 2 shows the flow diagram of the setup.
The mobile bed scrubber was a single stage pilot-scale unit.
It had a 45.7 cm (18 in.) square aluminum shell. The overall
height of the scrubber was 3 m (10 ft), allowing 1 m for the
fluidized bed plus 1.3 m of open column above the bed. Four flow
straightening vanes were installed below the bed. Six Plexiglas
windows, each of 15 cm x 40.6 cm, were installed on two adjacent
sides of the column so that the action of the packing and entrain-
ment flow pattern could be observed during operation. The packing
used for the bed was 3.8 cm (1.5 in.) diameter hollow polyethylene
spheres and the average weight of each sphere was 4.5 g. The
depth of the static packed bed was 25.4 cm (10 in.). The retaining
grids were made of hardware screen. The wire diameter was 0.16 cm
and the openings were 1.1 cm x 2.4 cm.
Air and water were used for study. Air flow rates to the
scrubber were measured through a pitot tube. The pitot tube
22
-------
ENTRAINMENT SEPARATOR
O ° i
r^ PTTDT L
BLOWER TUBE
AIR
MAKE-UP
WAlfcK "jl J
1
I
1
1
fc
r
o
O
o
o
0°0
1 jr
TV
P«IV
-------
was kept in one position after a complete traverse of the duct.
Water flow rates were monitored with a calibrated venturi meter.
Experimental Procedures
The purpose of the preliminary studies was to measure drop
size distribution and concentration of entrainment from a mobile
bed scrubber operating under fairly typical conditions. Air and
water flow rates covered common operating ranges, especially those
likely to cause heavy entrainment.
In starting the experiments, the water pump and blower were
turned on and the flow rates were adjusted to desired levels.
After steady conditions were obtained, entrainment drop samples
were taken at a location 76 cm (30 in.) above the top retaining
grids.
Drop Sampling Methods
Drop size distribution and concentration were measured with
a hot-wire anemometer and a cascade impactor.
Hot-Wire Anemometer -
The anemometer used in the study was the Model DC-1 Droplet
Counter manufactured by KLD Associates. It consists of an
electrically heated wire which is made of platinum and is 5 \im
in diameter and 1 mm in length. When the hot wire is placed in a
liquid aerosol stream it will display fluctuations in voltage due
to the cooling effects of the carrier stream and the liquid
droplets. Each drop contacting the wire cools a wire length
proportional to the drop diameter, reducing the wire electrical
resistance proportionately. The wire forms one leg of an electric
bridge which is imbalanced by this resistance change, thus
generating a pulse. An electronic circuit analyzes, counts, and
sorts the pulses into six channels or bins which have different
drop diameter intervals. The diameter interval for each bin
depends on the dc reference resistance network used. KLD calls
this resistance network a "Ladder Network." The DC-1 drop
counter is equipped with three "Ladder Networks." Table 3 shows
the diameter interval for each bin for these three "Ladders."
24
-------
TABLE 3. DROP DIAMETER INTERVAL FOR
DC-1 DROP COUNTER
Bin No.
1
2
3
4
5
6
Ladder #1
1 - 1.5
1. 5 - 2.25
2.25 - 3.4
3.4 - 5.0
5.0 - 7.5
> 7.5
Ladder #2
1 - 2
2 - 4
4 - 8
8 - 16
16 - 32
> 32
Ladder #3
1 - 3
3 - 9
9 - 27
27 - 81
81 - 243
> 243
Note: Diameters are in ym.
25
-------
The drop counter can detect the existence of a drop only
when the drop touches the hot wire (center of drop must be within
one drop radius of wire and attached to the hot wire). Thus, a
10 um diameter drop must be closer to the wire than a 100 ym
diameter drop in order to be detected. Therefore, when the hot
wire is placed in the gas stream, the gas volume sampled by the
hot wire differs for different drop sizes and is expressed by
the following formula:
Vs = UG t 1 (2 dd + dw) (12)
where d, = drop diameter, cm
d = hot wire diameter, cm
w '
1 = hot wire length, cm
t = sampling time, s
UG = gas velocity, cm/s
V = gas volume sampled, cm3
The drop counter only counts and sorts the drops into six
diameter intervals or bins. It does not give the exact size of
each individual drop. For this reason, we assumed that drops in
each bin had the same size and were equal to the arithmetic mean
diameter of the bin. The drop number concentration corresponding
to the bin is computed with the expression:
N.
ni = VT (13)
s i
where n- = drop concentration corresponding to ith bin,
#/cm3
Ni = number of drops counted in the ith bin, number
Vsi = gas volume sampled by the ith bin, cm3
"n^" is also the number of entrainment drops per cm3 in the
scrubber gas, which have diameters between the lower and upper
26
-------
diameter limit of the bin. The total drop number concentration
for all drop populations is the sum of all "n.'s"; i.e.
6
n =
where n = total drop number concentration, #/cm3
In obtaining a drop size distribution, the drop counter data
were treated as cascade impactor data with each bin correspond-
ing to a cascade impactor stage. The lower drop size limit of a
bin was assumed to be equivalent to the impactor stage cut diameter
For example, if Ladder #3 is used for measurement, the drop
number concentration or percent of total drops smaller than
3 ym; i.e. the lower limit of bin #2 is equal to the concen-
tration in bin #1. The concentration smaller than 9 ym or
the lower limit of bin #3 is the sum of concentrations in
bins #1 and 2. In general,
% by number smaller than d, .
x 100% (15)
n
where d , - = lower diameter limit of bin #i, cm
To obtain entrainment volume concentration, one simply multiplies
each bin's number concentration, "n." by the size drop volume cal
culated from the bin's average drop size. Total entrainment
volume is the sum of all bins. Cumulative entrainment volume
versus drop size can be obtained by the same method as the de-
rivation of cumulative number concentration.
27
-------
Cascade Impactor -
A few confirmatory measurements were performed with cascade
impactors. The sampling system arrangement is illustrated in
Figure 3.
Cascade impactors are usually used for size fractionation
of particles smaller than 20 um in diameter. Calvert et al.
(1977) reported that less than II of the mobile bed entrainment
has a diameter smaller than 10 ym. To prevent overloading the
impactor by big drops, a side port probe/pre-cutter was used
ahead of the impactor. The side port probe/pre-cutter was a
4.5 cm (1.75 in) I.D. Plexiglas tube with a 1.6 cm diameter
hole drilled on the wall. When pulling a sample through the
sampling system at a gas flow rate of 14 fc/min (0.5 CFM), the
pre-cutter theoretically removes all drops larger than 13 ym
in diameter.
Early efforts to detect fine drops by cascade impactors
were unsuccessful. It was found that fine drops evaporated
even under a nearly saturated gas environment. It ivas also
discovered that due to reduced pressure in the sampling sys-
tem, drops collected on impactor stage collection plates
evaporated. To alleviate this problem, sodium chloride was
added to the water.
The salt has the effect of reducing the vapor pressure
of the water and minimizing the evaporation of water from
drops as they leave the mobile bed and travel to the sampling
point.
The salt can also serve as a tracer. Once the drops are
collected by the impactor stages, evaporation would still
occur. If one assumes that evaporation does not occur before
collection by impactor stages, the amount of salt collected
on the impactor substrates allows the calculation of the
of the size distribution and concentration of drops existing
at the sampling point.
28
-------
tsJ
SIDE PORT
PROBE/
PRECUTTER
IMPACTOR
DRAIN
BOTTLE
VACUUM
ROTAMETER n GAUGE
THERMOMETER
GAS FLOW
OO 4
4
9 0
o
IMPINGERS
SILICA
GEL
VACUUM
PUMP
DRY GAS
METER
r
I
i
rl.
1
I
I
INCLINED
MANOMETER
Figure 3. Entrainment measurement sampling train.
-------
After the completion of a sampling run, the impactor sub-
strates were soaked in a known volume of distilled water. The
amount of sodium chloride leached out of the substrates was
then determined with a chloride specific ion electrode.
The drop size distribution was calculated from each impactor
stage catch, the cut diameter of the stage, and the sodium
chloride concentration in the scrubber liquid. Summation of
the pre-cutter catch and the impactor catch gives the entrain-
ment mass flow rate.
Entrainment Data
Preliminary Runs -
Several exploratory runs were conducted with the DC-1 drop
counter. It was found that due to the short sampling time
capability of the counter (99.9 seconds or 999 drops, which-
ever comes first) and the irregular nature of entrainment
generation, the agreement between runs was poor. However, if
the average of ten runs was used to compute the drop size dis-
tribution, results would be consistent. Therefore, in subse-
quent entrainment measurements each run consisted of at least
ten sets of data and the average of these was used for analysis.
Entrainment loadings were measured with the drop counter
at several locations in the same scrubber cross section. Ex-
cept at locations close to the scrubber wall, the entrainment
loadings at different locations were close to each other. In
subsequent measurements one point sampling was employed.
Seventeen runs corresponding to six different scrubber opera-
ting conditions were conducted to investigate whether there was
a discrepancy in the measurement of entrainment between the DC-1
drop counter and the cascade impactors. The scrubber operating
conditions are listed in Table 4. In all the runs, 12% by
weight of sodium chloride solution was used as scrubber liquid.
30
-------
TABLE 4 . PRELIMINARY ENTRAINMENT MEASUREMENT
SCRUBBER OPERATING CONDITIONS
Run
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Measurement
Device
Impactor
Impactor
Impactor
Impactor
Impactor
Impactor
Drop Counter
Drop Counter
Drop Counter
Drop Counter
Drop Counter
Drop Counter
Drop Counter
Drop Counter
Drop Counter
Drop Counter
Drop Counter
Superficial
Gas Velocity
(m/s)
2.9
2.9
2.9
2.9
2.9
3.4
2.9
2.9
2.9
2.9
2.9
2.9
3.4
3.4
3.4
3.4
3.4
Gas Flow
Rate
(m3/min)
36.8
36.8
36.8
36.8
36.8
42.5
36.8
36.8
36.8
36.8
36.8
36.8
42.5
42.5
42.5
42.5
42.5
Liquid Flow
Rate
(i/min)
254
254
322
322
394
299
254
254
322
322
394
394
299
299
369
369
460
Pressure
Drop
(cmW.C.)
4.6
4.6
4.8
4.8
5.8
5.2
4.6
4.6
5.1
5.1
5.1
5.1
5.1
5.1
6.. 4
6.4
7.4
Bed*
Expansion
(%)
110
110
110
120
130
130
110
110
120
120
130
130
140
140
160
160
190
H - H
Bed Expansion = — x 100%
H
31
-------
Since the drop counter's manufacturer, KLD Associates, claims
that suspended and dissolved solids will not affect the perfor-
mance of the counter, the usage of sodium chloride solution of-
fers an opportunity for checking the counter with the cascade
impactor.
Figure 4 shows the comparison of the entrainment volumes
measured by the drop counter and the cascade impactor. The
entrainment volume measured by the drop counter is much less
than that by the impactor. Cumulative entrainment volume
curves, calculated from cascade impactor data, show a tendency
to flatten out around 10 ymA. This is expected because of the
side port probe/pre-cutter which was used. The pre-cutter has
a theoretical cut diameter of about 13 ymA, which is close to
the first and second stage cut diameters of the A.P.T. cascade
impactor (13 ymA for the first stage and 9.8 ymA for the second
stage).
Another indication of the discrepancy between the two
measurement methods is the overall entrainment volume. The
DC-1 drop counter did not detect any drop larger than 243 ym
in diameter. Thus the overall entrainment volume measured
with the drop counter, is equal to the cumulative volume at
243 ym. For a superficial gas velocity of 2.9 m/s and a QL/QG
equal to 6.9 £/m3, the entrainment volume is 0.0245 ml/Nm3.
However, for the same condition, the volume based on pre-cutter
and impactor catch is 1.6 ml/Nm3, which is 65 times higher than
the drop counter result. Calvert, et al. (1977) reported some
entrainment data from a mobile bed scrubber. They used the
water balance technique to determine the volumetric entrain-
ment flow rate. At a superficial gas velocity of 2.9 m/s and
a QL/QG of 6.9 £/m3, the volumetric entrainment flow rate is
about 2 j,/m3-min, which corresponds to an entrainment loading
of 11.5 ml/Nm3. Calvert, et al.'s result is about 470 times
higher than the drop counter data.
32
-------
10
-2
10
-3
ID
_l
O
10"
10
-5
IMPACTOR DATA;
DROP COUNT
DATA
1. UG = 290 cm/s
QL/QG = 6.9
2. UQ = 290 cm/s
QL/QG = 8.7
3. UG = 340 cm/s
QL/QQ = 6.9
4. UG = 290 cm/s
Q./Qp = 10.8 £/m
LI" , , i i i i
10
DROP DIAMETER,
100
Figure 4. Comparison between cascade impactor and DC-1
drop counter data.
33
-------
The DC-1 counter manufacturer was questioned on these
discrepancies and it was found that they had the same experience.
Salt solution was atomized and both the cascade impactor and
the DC-1 drop counter were used to measure drop loading. They
found that the loading by the cascade impactor was about ten
times higher than by the DC-1.
Auxiliary Experiments -
In order to resolve the discrepancies between the two
sampling devices, some auxiliary experiments were carried out.
First, a small scale experiment was set up to duplicate the
findings. A 5 weight % solution of sodium chloride was atomized
with a Collison atomizer. Drop size distribution and loading
were then measured with the drop counter and the cascade impactor
simultaneously. The results confirmed the previous findings
that the drop size distributions measured with the two devices
were different.
It was suspected that even though evaporation of drops was
not present before the cascade impactor, evaporation did occur
inside the cascade impactor due to the pressure reduction from
stage to stage. To study the evaporation of drops within the
cascade impactor, the experimental apparatus shown in Figure 5
was built. A 5 weight % salt (NaCl) solution was atomized
with a Collison atomizer. The gas is mixed with dilution air
which passes through a conditioner which could either humidify
or dry and heat the dilution air. The aerosol was then passed
through an impactor to measure its size distribution. A
specific ion probe for Cl was used to determine the mass of
salt collected in these experiments.
The experiment consisted of two parts, and the first part
involved drying the atomized aerosol by mixing it with dehumidified
and heated dilution air. The size distribution obtained was that
of the dry residual salt particles. Observation of the sub-
strates indicated that no moisture was present for these runs.
Figure 6 shows the size distribution obtained for the dry
particle size.
34
-------
ATOMIZER
C. I.
MIXING
CHAMBER
DILUTION AIR
CONDITIONER
ROTAMETER
TO VACUUM PUMP
ROTAMETERS
FILTER
PRESSURE
REGULATOR
COMPRESSED
AIR
Figure 5. Sketch of experimental equipment.
-------
<
PS
w
E-
W
Q
,-J
E-"
Pi
8.0
6.0
4.0
2.0
1.0
0. 8
Q
§ 0.6
0.4
0. 2
I I
I I
5?6 SALT SOLUTION
d =1.7 ymA
Pg
a = 2.8
g
O
1 5 10 20 30 40 50 60 70 80 90
PERCENT BY WEIGHT UNDERSIZE, %
Figure 6. "Dry salt" particle size distribution.
36
-------
The size distribution of the parent drops can be obtained
from the size distribution of the residual salt particles since
the initial salt concentration is known. The initial drop dia-
meter is related to the salt particle diameter by the following
equation:
dd = dp (p/c) (16)
where: d, = drop diameter, cm
d = salt particle diameter, cm
p = salt density, 2.16 g/cm3
c = salt concentration in solution, g/cm3
This equation is good for diluted salt solutions where the den-
sity of the solution is close to that of pure water. The pre-
dicted drop size distribution from the atomizer is shown in
Figure 7. Note that the aerodynamic diameter is shown and that
conversion to physical diameter is required for application of
equation (16). Also shown on this figure are lines which repre-
sent the size distribution which would exist if a given fraction
of the water were evaporated from each of the drops. The upper
curve shows the parent drop distribution while the lowest curve
represents the measured dry size distribution.
The second part of the experiment involved measurement of
the wet drop size distribution. For this case the dilution air
was first saturated in a fritted disk bubbler so that particle
drying would not occur. The results of several sampling runs
with the University of Washington cascade impactor are shown
in Figure 8. As can be seen, the wet size distribution, in-
stead of being close to the predicted parent wet particle size
distribution, is close to the size distribution of dried salt
residues. This indicated that evaporation from particles occurred
Since the air was saturated with water vapor, evaporation from par-
ticles could not occur before the cascade impactor. Thus, it could
37
-------
H
W
<
i—i
Q
PL,
O
PARENT DROPS
501 EVAPORATION
75% EVAPORATION
DRY SALT - 100% EVAPORATION
10 20 40 60 80
PERCENT BY WEIGHT UNDERSIZE, %
Figure 7. Predicted drop size distribution as a
function of water evaporated.
38
-------
J.U
5
(METER, ymA
**i
i— i
o
w
U
HH
H ]_
PH
U
1 — 1
s
o 0.5
Pi
w
0.1
1 ' i i i i i i t \ i
: /o /
PREDICTED / d/fck
PARENT / nsVV/^
DROPS / ^tj—y
/ /
(At
/ /w
fXXi /
/^ , DRY SALT
: / y
/ GLASS IMPACTOR DATA
^ V' I
- y o
(I /2k U. OF W. IMPACTOR DATA
7 0
A
O
1 1 1 1 1 1 1 1 1 1
1 10 20 30 40 50 60 70 80 90 95
-
-
-
-
"™
-
-
98
PERCENT BY WEIGHT UNDERSIZED
Figure 8. Measured wet size distributions, U. of W.
impactor and series glass impactor.
39
-------
be deduced that evaporation occurred within the cascade impactor.
The reason for evaporation might be due to the pressure drop
across the jet plate and the negative pressure existing in the
cascade impactor.
To verify that evaporation did not occur before the cascade
impactor, a second series of experiments were conducted using a
series of glass impactors which allowed observation of the jet
and impaction plate. Figure 9 shows the experimental setup.
Each glass impactor consisted of one jet. By properly selecting
the jet diameter, the pressure drop across each stage could be
kept at a minimum.
A solution of 5% by weight salt was atomized and dried by
mixing it with heated dilution air. The dry salt size distri-
bution confirmed the previously obtained dry size distribution
from the University of Washington cascade impactor.
Several runs were then run without dilution air. The
results are shown in Figure 10 along with the predicted wet
particle size distribution. As can be seen, there is a reasonably
good fit between the glass impactor data and the predicted wet
particle size for diameters below 3 ymA. There is slight devia-
tion for drops larger than 3 ymA in diameter. The parent drop
size distribution was calculated from the dried salt distribution
measured with the University of Washington cascade impactor.
It is possible that the drying was incomplete in the first part
of the experiment, causing the predicted parent drop size to be
larger than it should be.
The wet size distribution measured with the University of
Washington impactor differs from that of the glass impactors
for particles smaller than 3 ymA in diameter. The agreement
between the two for larger particle sizes is fairly good.
Apparently, significant evaporation from the small particles
occurred within the University of Washington impactor. The
results shown in Figure 8 show that particles of 1 ymA in diameter
have their diameters reduced to 0.6-0.7 ymA by evaporation.
These experiments confirmed that evaporation from the
particles is due to the pressure reduction in the cascade
40
-------
ROTA-
METER
FILTER
PRESSURE Jy
M REGULATOR X
COMPRESSED
AIR
P T
TO
VACUUM
PUMP
FILTER
STAGE 3
ATOMIZER
Figure 9. Experimental setup for glass impactors.
-------
8 . U
6.0
*\
pi
E-
W
0 2.0
w
1— 1
H
Pi
PH
u 1.0
S
S 0.8
Q
0
w 0.6
0.4
0.2
1 l I I I 1 i 1 I i
/G52\ s*
O RUN 7/2L / ^ '
£ RUN 7/31 WET / r\
Q RUN 7/4L / ^ /
O RUN 7/5L / _yv /
(J RUN 7/8L - DRY // VJAK-> ^
/ /X/
PREDICTED PARENTS / /
\/ \
7 / \
\
/ / SALT PARTICLES
<4d 73
/x /
/ AVERAGE AP cm W.C.
/ WET DRY
APi. = 1.1 <1.0
AP2 =1.8 1.0
AP3 = 7.1 6.7
1 1 1 1 1 , ,
5 10 20 40 60 80
PERCENT BY WEIGHT UNDERSIZE, %
90 95
Figure 10. Wet size distribution obtained with glass
impactors.
42
-------
impactor. The pressure drop across the glass impactor was
relatively low. Therefore, little or no evaporation from
particles occurs and the measured wet particle size distribution
is close to predictions.
Entrainment Data Corrections -
By assuming evaporation from drops of 10% by weight of
sodium chloride was the same as that for 5% by weight, correc-
tions were made to the impactor data. The results are plotted
in Figure 11. As can be seen, there are still large discrepan-
cies between the two measurement methods. The remaining dis-
crepancies may be caused by the inability of the drop counter
to detect drops smaller than 1 ym in diameter.
These auxiliary experiments explained the differences
between the drop counter data and the cascade impactor data for
drops smaller than 10 ym in diameter. They did not answer the
question of why the overall entrainment loading measured with
the drop counter is much lower than that reported by Ca-lVert
et al. (1977) and that based on the pre-cutter catch of the
present study. It was discovered later that salt residues
coated the platinum wire used for drop detection after the drops
had been vaporized. The salt coating may reduce the sensitivity
of the platinum hot wire.
The scrubber liquid was subsequently changed to clean
water and a few entrainment measurements were carried out with
the drop counter. When using clean water, the total number
concentration of entrainment increased even though scrubber
operation was the same. For a superficial gas velocity of
2.9 m/s and Q /Qr of 6.9 £/m3, entrainment volume increased to
J.J (j
0.057 ml/Nm3 from 0.025 ml/Nm3. Even though the entrainment
loading detected was more than double that for salt solution,
it was still about 25 times less than that of pre-cutter catch
and about 200 times less than that reported by Calvert et al.
(1977) .
43
-------
10
-2
„ 10'
Lpl—l «
2 10
O
CORRECTED
IMPACTOR
DATA
10
- 5
1. UG = 290 cm/s
QL/QG = 6.9 Vm3
2. UG = 290 cm/s
QL/QG = 8.7 £/m3 -I
3. UG = 340 cm/s
QL/QG = 6.9 Jl/m3
4. UG = 290 cm/s
QL/QG = 10.8
| 1 i i I | |
10
DROP DIAMETER, ym
100
Figure 11
Comparison between cascade impactor and
DC-1 drop counter data.
44
-------
The drop counter also did not detect any drops larger than
245 ym diameter in the mobile bed depsite their being visible.
During drop measurements with the DC-1 drop counter, it was no-
ticed that some drops shattered upon impaction on the hot wire. There-
fore, the drop counter detected less drops than existed. This might
be the reason the drop counter gave a lower entrainment loading.
Goldschmidt and Householder (1969) have studied the measure-
ment of liquid drops with the hot wire anemometer. They gave
the following equation to estimate the largest drop size to hit
the wire without shattering.
12 a p
d
d max - —
UG
where d^ max = maximum drop diameter without shattering
upon impaction, cm
UG = gas velocity, cm/s
c?L = surface tension of liquid, dyne/cm
Py = liquid density, g/cm
. 3
For clean water, equation (17) becomes
d = Ml
d max 2
UG
(18)
This equation reveals that for a gas velocity of 2.9 m/s,
drops below 103 urn in diameter will never shatter upon impaction.
When gas velocity is increased to 3.8 m/s, the smallest drop to
shatter upon impact on wire is 60 ym in diameter.
Drop Counter Mobile Bed Entrainment Data
Since we are only interested in small drops, entrainment from
the mobile bed was measured with the counter for three levels of
gas velocity and five levels of clean water flow rate. Raw data
are given in Appendix A. Table 5 lists mobile bed operating
conditions for all runs.
45
-------
TABLE 5. DC-1 DROP COUNTER ENTRAINMENT DATA
AND SCRUBBER OPERATING CONDITIONS
Run
No
DC1
DC2
DC3
DC4
DCS
DC6
DC7
DCS
DC 9
DC10
DC11
DC12
DC13
DC14
DC15
DC16
DC17
DC18
DC19
Scrubber Operating Conditions
Superficial
Gas
Velocity
(m/s)
3.5
3.5
3.5
3.5
3.5
3.5
3.5
3.5
3.4
3.4
2.9
2.9
2.9
2.9
2.9
2.9
2.9
2.9
2.4
Gas Flow
Rate
(m3/min)
44.2
44.2
44.2
44.2
44.2
44.2
44.2
44.2
42.5
42.5
36.3
36.3
36.3
36.3
36.3
36.3
36.3
36.3
29.7
Liquid
Flow
Rate
(£/min)
250
250
318
318
356
356
386
386
454
454
454
454
390
390
318
318
265
265
265
Pressure
Drop
(cm W.C.)
5.1
5.1
5.3
5.3
5.8
5.8
6.1
6.1
6.4
6.4
6.1
6.1
5.6
5.3
5.3
5.3
5.1
5.1
5.1
Bed
Expansion
(%)
140
140
160
160
180
180
190
190
220
220
170
170
140
140
105
100
90
90
50
Liquid Entrainment
Entrainment
Flow Rate
(ml/m2-min)
35.5
-
-
42
55
-
-
45.8
-
82.6
83.5
-
-
23.5
13.9
-
-
9.6
4.6
Mass Median
Drop Dia.
(urn)
330
198
176
209
156
174
217
141
298
160
382
174
105
127
100
137
224
93
54
Geometric
Standard
Deviation
3.2
3.0
3.0
3.0
2.9
2.9
3.0
2.9
3.1
2.9
3.2
2.9
2.6
2.8
2.7
2.8
3.0
2.7
2.4
-------
TABLE 5. DC-1 DROP COUNTER ENTRAINMENT DATA
AND SCRUBBER OPERATING CONDITIONS
(continued)
Run
No
DC20
DC21
DC22
DC23
DC24
DC25
DC26
DC27
DC28
Scrubber Operating Conditions
Superficial
Gas
Velocity
CWs)
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
Gas Flow
Rate
(m3/min)
29.7
29.7
29.7
29.7
29.7
29.7
29.7
29.7
29.7
Liquid
Flow
Rate
(Vmin)
265
318
318
356
356
390
390
454
454
Pressure
Drop
(cm W.C.)
5.1
5.1
5.1
5.3
5.1
5.8
5.8
6.1
6.1
Bed
Expansion
C%)
60
70
60
80
80
90
90
120
120
Liquid Entrainment
Entrainment
Flow Rate
(ml/m2-min)
-
-
8.1
12.5
-
-
10.5
11.2
-
Mass Median
Drop Dia.
(um)
198
161
149
114
135
168
106
80
162
Geometric
Standard
Deviation
2.9
3.0
3.0
2.7
2.7
2.9
2.7
2.6
2.9
-------
The size distribution measured by the drop counter is
an approximately log-normal number distribution. For prac-
tical usage, it is more convenient to express the size
distribution based on mass instead of number. Since for a
log-normal distribution the number and mass distributions
have the same geometric standard deviation, the following
equation is used to convert number median to mass median
diameter (see Orr, 1966):
d
In -^ = 3 In2a (19!
ddN
where d,,, = mass median drop diameter, ym or cm
d-,N = number median drop diameter, ym or cm
a = geometric standard deviation, dimensionless
&
The calculated mass median drop diameter and geometric
standard deviation are tabulated in Table 5. Figure 12 is a
plot of mass median drop diameter versus liquid/gas ratio,
with superficial gas velocity as a parameter. It can be seen
that under normal industrial scrubber operating conditions, the
liquid entrainment has a mass median drop diameter of approxi-
mately 170 ym, which is smaller than that reported by Calvert
et al. (1977). Calvert et al. (1977) used chemically treated
filter papers to determine the drop size distribution. The
mass median drop diameter reported by them is approximately
400 ym.
The geometric standard deviation for all drop counter test
runs averages about 2.9.
Figures 13 through 15 are cumulative entrainment volume
plots for the three gas velocities studied. These plots can
be useful for designing and specifying entrainment separators.
The dashed lines in the figures are the calculated maximum drop
size without being shattered upon impaction on the hot wire.
48
-------
600
500
1
e*T
H 400
w
i— i
o
o 300
g 200
CO
CO
100
0
(
i i i I i i i i r T i i i i i
/\ UG = 240 cm/s
Aur = 290 cm/s
u —
/-v UG = 340 cm/s
/~*\ uf = 350 cm/s
A
O
O
0 * °>
(A ^ O ^
A A A <^> O
O ^
i I 1 1 1 1 1 1 1 1 1 1 i 1 1
3 2 4 6 8 10 12 14 1<
LIQUID TO GAS RATIO, £/m3
Figure 12. Mass median drop diameter of entrainment from mobile bed
scrubber (Drop counter data).
-------
500
CD
o
100 I-
w
H
U4
PH
O
Pi
un = 240 cm/s
u
QL/QG (cm3/cm3)
1-1 A-* i r\ — ^
0.1
CUMULATIVE ENTRAINMENT VOLUME, ml/Nm3
Figure 13. Cumulative entrainment loading (Drop counter data).
-------
500
100
s
3.
. 50
O
g 10
ur = 290 cm/s
u
I I I I I I I I
I I I I Mill
d ,max
j L
l I
QL/Q (cm3/cm3)
10
- 5
10
10"3 0.01
CUMULATIVE ENTRAINMENT VOLUME, ml/Nm3
0.1
0.4
Figure 14. Cumulative entrainment loading (Drop counter data)
-------
Cn
K)
500
100
^ 50
*v
OS
w
H
W
<
i—i
Q
PH
§ 10
10
- 5
i i I
ur = 350 cm/s
b
d,max —
I I I I I I I i I
l I I I i I I I
i I I i I i i
i i i i il i i
10
. * 3
10 J 0.01
CUMULATIVE ENTRAINMENT VOLUME, ml/Mm3
0.1
i i i
1.0
Figure 15. Cumulative entrainment loading (Drop counter data).
-------
Figure 16 is a plot of total entrainment flow rates versus
liquid/gas ratio with superficial gas velocity as parameter.
It can be seen that entrainment flow rate is both a function of
gas flow rate and liquid flow rate. The entrainment flow rate
may be low due to shattering of drops. Calvert et al. (1977)
gave a similar plot which has much higher entrainment loadings.
53
-------
100
•H
e
80
60
40
20
1 1 r
I I
290 cm/s
240 cm/s
024 6 8 10 12 14 16
QL/QG x 103, cm3/cm3
Figure 16. Entrainment floiv rate (Drop
counter data).
54
-------
SECTION 4
MOBILE BED F/C SCRUBBER PILOT PLANT
MOBILE BED F/C SCRUBBER PILOT PLANT
The schematic flow diagram of the mobile bed F/C scrubber
system is shown in Figure 17. Components of the scrubber system
are listed in Table 6. Table 7 describes the flow rates in the
lines shown in Figure 17 when the inlet gas stream to the scrub-
ber is 28 Am3/min (i^QOO ACFM) and saturated with water vapor at
60°C.
Figures 18 through 22 are detailed design and layout of the
mobile bed F/C scrubber pilot plant. A brief description of the
pilot plant components are given below.
Mobile Bed Scrubber
The mobile bed scrubber shell was made from fiberglass rein-
forced plastic duct with an inside diameter of 0.51 m (20 in.).
The overall height of the scrubber was 6.1m (20 ft) of which
2.75 m (9 ft) was reserved for mobile beds. A maximum of three
stages can be housed in this space. There was an empty space of
0.92 m (3 ft) above the top retaining grid. This space was pro-
vided for the settling out of large entrainment drops. Figure 23
is a schematic diagram of the mobile bed scrubber.
Mobile Bed Packing -
The mobile bed packings were hollow polypropylene spheres.
Three sphere sizes -- 2.5 cm, 3.8 cm, and 5.1 cm diameter --
were studied. The average weight for each sphere was 1.5 g,
4.5 g, and 8.0 g for 2.5 cm. 3.8 cm, and 5.1 cm diameter balls,
respectively.
Retaining and Supporting grids -
Two types of bed supporting and retaining grids were used
in the mobile bed scrubber. They were hard\vare screens and
55
-------
* * **
IB!
D
11
K
\
10
Figure 17. Process flow diagram of mobile bed F/C scrubber system.
-------
TABLE 6. LIST OF PILOT PLANT COMPONENTS
EQUIPMENT
A. Gas Burner
Nozzle mix type, natural gas open flame gas burner
with a capacity of 3.52 x 105 J/s (1,200,000 Btu/hr).
B. Furnace
Insulating fire-brick lined, rectangular box with
provision for dilution air and aerosol inlet.
C. Quencher
0.76 m in diameter and 2.44 m long horizontal vessel
with fine water spray.
D. FF/C Scrubber
Three stage mobile bed.
E. Entrainment Separator
Staggered tube bank type demister.
F. Water Cooling Tower
A horizontal spray type. Cooling range 17°C (30°F)
at 340 H/min (90 GPM).
G. Blower
Centrifugal fan with capacity of 1.23 m3/sec (2,610 CFM)
and static head of 5.2 kPa (53 cm W.C.)
H. Cooling Tower Fan
9.4 m3/s and 0.5 kPa (5 cm W.C.) head.
I. Pump
75 fc/min and 350 kPa pressure (20 GPM and 120 ft head).
J. Pump
680 £/min and 118 kPa pressure (180 GPM and 40 ft head).
K. Pump
610 H/min and 350 kPa pressure (160 GPM and 120 ft head)
57
-------
TABLE 7. FLOW RATES AND CONDITIONS FOR MOBILE
BED F/C SCRUBBER PILOT PLANT
Stream
No.
1
2
3
4
5
6
7
8
9
10
11
Composition
Natural Gas
Room air containing
0.007 g FhO/g dry air
Room air containing
0.007 g H20/g dry air
Test aerosol dust
Water
Aerosol - air mixture
containing 0.15
g H20/g dry air
Air mixture contain-
ing 0.043 g H20/g
dry air
Water
Water
Humid air
Water
Temp.
°C
20
29
29
32
60
38
32
49
30
49
Gauge Press.
cm W.C.
+17.8
305
-17
-36
305
+ 5
Volume
m3/s
0.0063
0.069
0.28
0.47
0.44
708
Flow
l/s
1.25
3.5
3.5
—
58
-------
AIR POLLUTION TECHNOLOGY,
Inc.
Figure 18. Mobile bed layout.
-------
1
T
V
l
_
V
!
"i
i
r-
aV
nV
,11V
/
surrow TffAr
'1
_. ^?
*>x
< At-lAi QGftQH
-JOUKO-PYNE"
"?4fJE KA'L
sse sountoEs-r e#
^jiiUN/UUM
-- Wfi.0 4L.lt,
BOTTOM Of
^-COM
MOS/LE BE.D ELEVATION
Figure 19,
INTERNAL SUPPORT
Mobile bed and entrainment separator
MR POLLUTION TECHHOLOQV^
MO&LE RED AND
SEPARATOR
-------
4
^
LEFT ELEVATION
RIGHT E.LBVATIOU
FRONT ELE/ATIOH
—SUMP SIDES 4. BOTTOM
rf Sin PLYmaP f/OEG-
'JO/KS CORNERS
' BOTH SUMPS
i— t rp rn
r
»r
~-
[
(
f-eS .
\
-)— '
— a*& FBtlL p/TAttf
t_l LJ L
.r-StW OOTLJMS (TV?}
\ — i i — i ^ • r~i
n
LX-—
1
^^_ TO
PLAM
SECTION (B\
SECTION
Figure 20. Cooling tower framing and mobile bed sumps.
AIR POLLUTION TECHNOLOGY,
COOLING TOWER FRAMING
CT 4 M- a SUMPS
-------
MR POLLUTION TECHNOLOGY,
Inc. c.m.™
BLOWER VIBRATION MOUNTS
' *U? 11 ^. W
'"r<"«'' ,r^v<-d] Nr- V
f^fVMA -mUEK FXtJ
" * • /'-e
Figure 21. Quencher, furnace and blower vibration mounts.
-------
PKKEMS
V
WEST
MQ&ILE B£D PLATFORM AND ELEVATIONS
. j',.tfSlVVflt's%"s%." ?".._ j'-/"
PK*y NO,3 IS taPrrsET to" LSFT
fHORMAL. PVSn/OH, SFKAY MtNlfVLO*
TOWER AKt ON $'-(,*
OOOo
/•/ /-Z W 3-1
OOOo
AJ /-* AS 3-Z
•OOOo ^^
/-£ 1-L £-3 r-S ^ f.t s.3
FRONT
CONTROL
UWSTKJT UtlHSfK
COOLING TOWBR PIPING
SECTION
A-A
Figure 22. Mobile bed platform and cooling tower piping elevations.
AIR POLLUTION TECHNOLOGY,
5U Dieio - IIlC« CillfomlM
MOBILE BED PLATFORM &
COOLING TOWER PIPIN6 ELB/
PREPARED W»
JTOJ. EMGII^ 5m~
-------
K
SCRUBBER LIQUID INLET
7.62 cm (3" sch. 40)
u
OO
BALL RETAINING
GRID
GAS INLET
". 6
oo u
CTl
. e
r—I
CT)
CTl
I '
I ,
CT>
E
o
to
S
u
Figure 23. Mobile bed scrubber
64
-------
plastic nets. The wire diameter of the hardware screen was
0.16 cm (1/16 in.) and the openings were 1.1 cm x 2.4 cm (7/16
in. x 15/16 in.). The open area was about 82%. The plastic net
was made of polypropylene ribbons which were 0.40 cm (5/32 in.)
wide. The openings were 0.6 cm x 0.75 cm (15/64 in. x 19/64 in.)
and the open area was about 40%. The grids were rested on square
frames made of 2.5 cm aluminum angles 33 cm on a side. The effec-
tive open areas of the hardware screen support and the plastic
net support were 681 and 34%, respectively.
Liquid Distribution -
Scrubber liquid was distributed evenly to the mobile bed
just above the third stage top retaining grid through four PVC
pipes (1-1/2" sch 40) drilled with 1.27 cm (0.5 in.) holes. The
inlet header pipe was 3" sch 40.
Entrainment Separator
The entrainment separator consisted of six rows of staggered
PVC pipe banks. The external diameter of the pipe was 2.13 cm
( 1/2" sch 40). The spacing between pipes within a row was 3.4
cm center to center. The cross section of the entrainment
separator was 34.3 cm square.
Cooling Tower
The cooling tower was a horizontal cocurrent spray
chamber. The cross section was 1.53 cm x 2.14 cm (5 ft x
7 ft) and the length of the spray section was 5.8 m
(19 ft).
There were two entrainment separators at the outlet of the
spray section. The first entrainment separator was zigzag baf-
fles which were made of corrugated fiberglass. The second en-
trainment separator was pipe banks of the same design as that
for the mobile bed scrubber. The baffles were used to remove large
drops and to reduce the entrainment loadings reaching the pipe
bank entrainment separator.
65
-------
There were five spray banks, each with twenty spray
nozzles. The liquid flow rate was 75 £/min per spray bank
at a pressure of 370 kPa (40 psig).
Ambient air was induced through the cooling tower by a
vane-axial fan (Chicago Blower vane-axial fan size 40-1/4) with
a capacity of 9.4 m'/s (20,000 CFM) at a pressure of 0.5 kPa
(5 cm W.C.) and a 15 hp motor.
Quencher
The quencher was a horizontal vessel with fine water spray.
The spraying rate was about 75 fc/min (20 GPM) . The vessel was
0.76 m (2.5 ft) in diameter and 2.44 m (8 ft) in length.
Furnace
The furnace was a rectangular box lined with insulating fire
brick. The dimensions of the furnace were 0.92 m wide, 0.92 m
tall, and 1.53 m long (3' x 3' x 5'). There were fire brick baf-
fles in the furnace to induce turbulence and to give better mix-
ing of combustion flue gas and air.
The gas burner was a nozzle mix type natural gas burner
(Hauck NMG-130). The capacity of the burner was 3.52 x 105 J/s
(1.2 x 106 BTU/hr). Some safety devices were installed to en-
sure the proper operation of the burner. The devices included
a UV scanner (Fireye scanner UV-1) to sense the pilot flame, a
temperature controller at the quencher outlet to guard against
malfunction of the quencher, a pressure switch in the combustion
air line to ensure a consistent air supply to the gas burner,
high and low pressure switches in the natural gas pipe line to
guard against abrupt changes in natural gas pressure, and two nor-
mally closed solenoid valves in the main and pilot gas line.
A control system (Fireye chasis UVM-2A) was installed to co-
ordinate the functionings of the scanner, pressure switches,
temperature controller, and to regulate the solenoid
valve.
66
-------
Gas and Liquid Mover
Air and water were used for studies. Air was induced
through the mobile bed by means of a centrifugal fan (Chicago
Blower Turbo-Pressure blower 2T-15-12) with a capacity of 1.23
m3/s (2,610 CFM) and a static pressure of 5.2 kPa (53 cm W.C.)
Two rectangular tanks with dimensions of 84 x 120 x 210 cm
were used to store and supply the scrubber liquid. One 55 gal-
lon barrel was used to serve as the scrubber sump and water seal.
Scrubber liquid was pumped to the top of the mobile bed by means
of a centrifugal pump (Allis Chalmers C-l) . The capacity of the
pump was 680 £/min at a pressure of 118 kPa (180 GPM at 40 ft
head).
The cooling tower pump was a centrifugal pump manufactured
by Allis Chalmers with a capacity of 610 £/min at a pressure of
350 kPa (160 GPM at 120 ft head). The quencher pump was a small
centrifugal pump with a capacity of 75 £/min at a pressure of
350 kPa (20 GPM at 120 ft head).
INSTRUMENTATION AND CALIBRATION
The gas flow rate through the mobile bed scrubber was moni-
tored by a venturi meter located in the scrubber outlet duct.
It was calibrated against flow rates measured by standard pitot
tube traverses.
Liquid flow rates were also measured by venturi meters.
They were calibrated by measuring the volume of liquid flowing
through the pipes in a given time.
Temperatures in the scrubber system were measured by copper-
constantan (type T) thermocouples. The thermoelectric voltages
were recorded on a strip chart recorder equipped with a potentio-
metric amplifier. The thermocouples were calibrated against a
standard mercury bulb thermometer using constant temperature
baths. In the latter part of the experiment, type K thermocouples
and a digital temperature readout (Omega Engineering model 175-
KF1) were used.
67
-------
The scrubber gas line pressures were measured by "Magne-
helic" pressure gauges. Pressure differences in the liquid
venturi meters were measured with well-type mercury manometers.
Pressure difference in the gas venturi meter was monitored with
an inclined manometer.
Moisture contents in the inlet and outlet gas streams of
the scrubber were measured by wet and dry bulb thermometers.
PARTICLE GENERATOR
The schematic diagram of the particle generator is shown
in Figure 24. The aerosol was generated by redispersing the
powder with a jet of compressed air which was ionized by passing
through a Polonium 210 ionizing air nozzle (3M Company, model
906) .
In the beginning of a test run, aerosol powder was loaded
into the vibrating hopper and was fed by a screw feeder arrange-
ment into the compressed air jet. The feed rate of the powder
was controlled by regulating the screw turning speed through a
variable speed motor. The air jet dispersed the powder into
airborne particles. The dispersed aerosol next entered the
cyclone pre-cutter to remove coarse particles from the dispersed
aerosols. The aerosol then passed through a parallel plate
electrostatic precipitator to remove any charged particles be-
fore entering the scrubber inlet duct.
The electrostatic precipitator consisted of 19 alternatively
charged flat plates. The plates were 30.5 x 91.5 cm 20 gauge
aluminum plates. Plate spacing was 1.5 cm. The maximum applied
voltage was 10 kV.
PARTICULATE SAMPLING SYSTEM
Two identical, modified EPA Method 5 particle sampling trains
with in-stack University of Washington Mark III cascade impactors
were used to measure the particulate loadings and size distri-
butions in the scrubber gas inlet and outlet streams simultan-
eously. Isokinetic sampling technique was used in the entire
experimental program. The impactors in the inlet and outlet
68
-------
MOTOR
VIBRATING
HOPPER
COMPRESSED
AIR
CHARGE
NEUTRALIZER
13-
AEROSOL
TO SCRUBBER
INLET DUCT
POLONIUM 210
IONIZING AIR
NOZZLE
CYCLONE
PRECUTTER
Figure 24. Powder redispersion particle generator.
69
-------
were allowed to heat up to gas temperature before samples were
taken.
The particle number concentration was measured using a
batch dilution system and two particle counters - a Gardner Con-
densation Nuclei Counter (CNC) and an Electrical Mobility Ana-
lyzer (EMA) by Thermal Systems.
DATA REDUCTION METHODS
Particle Size Distribution
The particle size distributions were measured gravimetric-
ally using the cascade impactor data.
The particle diameter measured by an impactor is called
"aerodynamic diameter" and it has the units of "aerodynamic
microns, ymA." This is the effective diameter for particle
separation by inertial impaction and it takes into account the
effects of particle density and particle slip between gas mole-
cules. Aerodynamic diameter is related to geometric diameter
(actual size) by the following relationship:
dpa = dp
-------
Cumulative mass of particles collected on a 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 determined from the sampling flow rate
and impactor calibration.
The particles were assumed to have a log-normal distribution,
The cascade impactor stage cut diameters were plotted against the
percent by weight of particles smaller than the cut diameters on
a log-probability paper. A straight line was fitted to the data
points. The 50% value of "d " is the geometric mass median dia-
pa
meter, "d " and the 84.1% value of "d " divided by the mass
median diameter is equal to the geometric standard deviation,
"V-
Particle Loadings and Overall Penetrations
The total loadings in the inlet and outlet ducts were cal-
culated in the following manner:
1. The sample flow rate was converted to the standard
conditions of 0°C and 76 cm of mercury pressure.
2. Total weight gain on the sampling elements was
measured with an analytical balance, Sartorius
Model 244, +0.05 mg precision.
3. The particle mass loading, c (g/DNm3), was cal-
culated from:
(Total weight gain, g)
c
(Sampling rate,DNm3/min)x(Sampling time,min)
4. The overall penetration was calculated from:
Ft = . (22)
Si
where "c " and "c -" were the outlet and inlet
po pi
particle loadings measured simultaneously for the
run.
71
-------
The computation of penetration as a function of particle
aerodynamic diameter, or the grade penetration through the
scrubber, was done by a stepwise graphical procedure. The pro
cedure is based on the following equations:
Overall penetration can be defined as:
Pt = — -
Pt-, dc
d p
(23)
where "c " is the total particle loading and "Pt," is the pene-
tration for particle diameter "d " and it is given by:
pa
pta •
(dpa)
o
dc
(V)
(24)
•J 1
where
dc
Ka)
is the slope of cumulative mass loading less
pa;
versus the aerodynamic particle diameter curve at
than "d "
pa
"d a", and equals "f
Thus, to determine the fractional penetration, the follow-
ing procedure was followed:
1. Cumulative mass loading for all the stages and the
filter, below the stage with a cut diameter of
"d ", was plotted against "d " from the inlet
and outlet cascade impactor samples.
2. Slopes of the inlet and outlet plots above were
determined for several "d " values in the range
Pa
of 0.4 to 10 ymA. The fractional penetrations
were then determined from the ratio of the
slopes, as described above.
72
-------
3. The fractional penetrations were plotted against
particle diameters. The curve is termed grade
penetration curve. The diameter whose penetra-
tion is 0.5 is the performance cut diameter of
the scrubber.
Particle Number Concentration
Particle number concentration is an important parameter for
F/C scrubbing. In the present study, particle number concentra-
tion was measured by means of a Gardner Condensation Nuclei
Counter (C.N.C.) and. a Thermal Systems Electrical Mobility
Analyzer (E.M.A.).
Particle number can also be calculated from cascade impactor
data if the size distribution is log-normal. The procedures are
as follows:
1. Convert the impactor stage aerodynamic cut dia-
meter into physical size diameter (equation 20).
2. Plot the physical cut diameter against the
percent by weight of particles smaller than
the cut diameter on a log-probability paper.
3. Obtain the physical geometric mean diameter,
d ' , and geometric standard deviation, a ',
r o &
from the plot.
4. Calculate the mass mean diameter by the follow-
ing equation:
lndm = lndp'g - 1.5 lnVg (25)
where d = mass mean diameter, urn
m
d ' = physical geometric mass median diameter, ym
a ' = physical size geometric standard deviation,
o
dimensionless
5. Number concentration is calculated from
n = A = 6JL_ ± x 10i2 (26)
p V TT p d 3 V
s p m s
73
-------
where n = particle number concentration, #/cm
N = total number of particles, #
V = volume of gas sampled, cm
3
s
m = total mass of particles, g
p = particle density, g/cm3
74
-------
SECTION 5
EXPERIMENTS
EXPERIMENTAL CONDITIONS STUDIED
The mobile bed scrubber performance was evaluated for three
different scrubber operation modes. They were: 1) cold, 2)
slurry scrubbing, and 3) F/C scrubbing.
Cold Runs
A cold run is defined as the condition in which both the
gas and liquid going to and from the scrubber are at ambient
temperatures. No preconditioning is applied to either one.
This operation mode was designed to isolate the collection
mechanisms which are responsible for particle collection in
a mobile bed.
Air and water were used for this operation mode. Variables
studied included: superficial gas velocity, liquid flow rate,
packing diameter, static bed height, types of supporting grid,
types of aerosol, and number of mobile bed stages.
The ranges of the variables studied were:
1. Superficial gas velocities ranged from 200 cm to
500 cm.
2. Liquid flow rates ranged from 136 5,/min to 410 £/min.
3. Packing diameters. Three types of packing were
studied. They were 2.5, 3.8, and 5.1 cm diameter
polypropylene spheres. The average weight for each
sphere was 1.5 g, 4.5 g, and 8.0 g for 2.5 cm, 3.8
cm, and 5.1 cm diameter balls, respectively. Since
the 3.8 cm diameter balls are commonly used in in-
dustrial mobile beds, most of the experiments were
conducted with this size packing.
75
-------
4. Static bed height of the packing. Three bed depths
were studied. They were 15, 23, and 30 cm.
5. Number of mobile bed stages. One, two, and three
stage mobile beds were studied.
6. Supporting grids. Two types of supporting and
retaining grids were studied. They were hard-
ware screen and plastic nets. The characteris-
tics of the supporting grids were presented in
the last section.
7. Aerosol. Four types of aerosol dust were used.
They were titanium dioxide, red iron oxide,
power plant fly ash, and gray-iron cupola dust.
Slurry Scrubbing
Experience with the SO scrubber systems at many power
jC
plants and experimental facilities has shown that the entrain-
ment of scrubber liquid is a. significant source of particulate
effluent. The main objectives of the slurry test were:
1. Characterize the effect of scrubber slurry on
emissions due to entrainment.
2. Evaluate any influence on the particle collection
mechanism due to the slurry properties.
The experimental system for the slurry study was the same as
the setup for the cold operation mode experiment, except for
the addition of CaC03 powder to the scrubbing liquor. No other
chemical was added to modify the slurry. The slurry pH value
and density were determined by an Electro-Mark pH Analyzer and
by weight-volume measurement, respectively. The conditions
studied were as follows:
1. Mobile bed scrubber
A three stage mobile bed, each packed with 3.8 cm
diameter spheres to a depth of 23 cm, was studied.
2. CaC03 concentration (by weight)
5% (pH value = 8.00, density = 1.07 g/cm3) and 10% (pH
value = 8.2, density = 1.18 g/cm3)
76
-------
3. Superficial gas velocity
250 and 320 cm/s of air velocity were tested.
4. Liquid flow rate
227 £/min and 318 £/min were studied.
5. Aerosol powder
Gray iron cupola .dust was used as the aerosol powder.
F/C Scrubbing
F/C (flux force/condensation) scrubbing is a particle col-
lection mechanism which combines the collection phenomena of
diffusiophoresis, thermophoresis, and particle growth by con-
densing water vapor on the particles. Depending on the conden-
sation ratio, F/C scrubbing can be very efficient in removing
submicron particles.
The benefit of condensation scrubbing was studied under
this operation mode. The study was focused on the effect
of condensation ratio and the particle collection effic-
iency .
The following conditions were studied:
1. Packing - 3.8 cm diameter hollow polypropylene
spheres.
2. Bed height - 23 cm static bed height.
3. Gas velocity - 210 and 340 cm/s
4. Liquid flow rate - 273 £/min.
5. Condensation ratio - 0 to 0.25 gram of water
condensed per gram of dry gas scrubbed.
6. Aerosol - gray iron cupola dust.
7. Supporting grids - hardware screen and plastic
net.
EXPERIMENTAL PROCEDURES
Cold Runs
The mobile bed scrubber was the only equipment in operation
for this operation mode. The cooling tower, burner, and the
quencher were idle. Air and water at ambient conditions were
used for study.
77
-------
In starting the experiment, air and water flow rates were
adjusted to and maintained at the desired levels. After steady
conditions were reached, gas and liquid temperatures and the pres-
sure drop were recorded. Aerosols were injected into the system
at a location either before or after the quencher. Particle size
distribution and loading were measured simultaneously at the
scrubber inlet and outlet ducts by means of in-stack cascade
impactors. Performance of the scrubber was calculated from the
impactor data.
Slurry Scrubbing
The experimental procedures were the same as that for cold
runs except slurry was used as the scrubber liquid.
F/C Scrubbing
Clean water and air were used for study in this mode of
operation. In starting a run, the following procedures were
followed:
1. Start the scrubber liquid pump and adjust the
flow rate to desired level.
2. Start the quencher and adjust the flow rate.
3. Start the cooling tower and adjust spraying
rate.
4. Turn on the mobile bed blower and adjust the
flow rate.
5. Turn on the burner, monitor the gas temperature
at the furnace outlet, and adjust the burner to
obtain the desired gas temperature.
6. Readjust the flow rate through the mobile bed.
7. After steady state condition is reached, re-
cord the gas temperatures, liquid temperatures
and pressure drops.
8. Inject aerosols into the furnace and start the
samp1 ing equipment.
78
-------
EXPERIMENTAL RESULTS
Experimental and sampling procedures, and the methods of
data analyses and calculation of results are described in the
preceding sections. During the experimental study, scrubber
performance was determined as fractional penetration of par-
ticles (with respect to the aerodynamic particle diameter) and
the overall particle penetration through the scrubbers. Since
the scrubber inlet particle characteristics (size distribution
and number concentration) were different for each run, the
fractional penetrations provide a common base for comparing
scrubber performances for different conditions.
The scrubber operating conditions and performance are
tabulated in Appendix B. The fractional penetration plots for
all cascade impactor runs are given in Appendices "C"
through "E".
DISCUSSIONS
Cold Runs
Pressure Drop -
A typical diagram of the pressure drop across a mobile
bed is shown in Figure 25 (Tichy and Douglas, 1973). If the
liquid velocity is kept constant while the gas velocity is
increased from zero, the following sequence of phenomena can be
observed. At first, the packing spheres remain motionless in a
static condition. The pressure drop in this region increases
sharply with an increase in gas velocity. The condition continues
until the gas velocity reaches the minimum fluidization velocity;
i.e., the gas velocity at which the pressure drop is equal to the
weight of packing plus weight of liquid holdup.
Increasing the gas velocity beyond the minimum fluidization
velocity results in expansion and fluidization of the bed. The
bed behaves like an aggregated fluidization bed and is called
a mobile bed. For a mobile bed operating in the mode of
79
-------
LOCUS OF
MAX. FLUIDIZATION
VELOCITY
u
L2
ex
OPERATING REGION
LOCUS OF
MIN. FLUIDIZATION
VELOCITY
REGION OF CONSTANT
LIQUID HOLDUP
REGION
OF
INCREASING
LIQUID HOLD-
UP ^
Figure 25.
Pressure drop across one stage of
a mobile bed versus gas velocity
with liquid velocity as parameter
80
-------
fluidization without flooding (see next section for definition),
liquid holdup at a fixed liquid rate remains approximately con-
stant for different gas rates. The pressure drop across the bed
levels off to a constant value as shown in Figure 25.
As the gas velocity increases further, approaching the ter-
minal settling velocity of the wetted packing spheres, liquid
holdup gradually increases. The pressure drop across the bed
increases accordingly.
When the gas velocity reaches the terminal settling velocity
of the wetted packing, the packing is pushed up underneath the
upper retaining grids. When this happens, the mobile bed be-
comes a packed bed and pressure drop again increases sharply
with gas flow rate.
The above sequency was confirmed in the present study. The
experimental pressure drop curve for a mobile bed with hardware
screen support (Figure 26) is similar to Figure 25.
For a fully developed mobile bed, the pressure drop is made
up of the sum of the pressure drops due to the dry retaining
grids, the weight of dry packing, the liquid layer retained on
the supporting grid, the liquid retained in the bed, atomization
of liquid drops and the wall friction (Blyakher et al., 1967;
Kito et al., 1976); i.e.
AP = AP, + APT, + APT + AP + AP + AP. (27)
wbLhLacf ^
where: AP = overall pressure drop for one stage of the
mobile bed, cm W.C.
AP, = pressure drop due to weight of dry packing, cm W.C.
APT, = pressure drop due to liquid holdup, cm W.C.
APT = pressure drop due to liquid head retained on
±j
supporting grid, cm W.C.
AP = pressure drop due to drop atomization, cm W.C.
cl
AP = pressure drop due to wall friction, cm W.C.
AP_p = pressure drop due to friction of supporting
grid, cm W.C.
If there is no liquid retained on the supporting grids and the
losses due to wall friction, supporting grid friction, and drop
81
-------
atomization are small, the pressure drop across the mobile bed can be
considered equal to the sum of the weight of the packing and the liquid
holdup in the bed (Tichy and Douglas, 1973; Kito et al. , 1976), i.e.
=
-------
ex
f~-:
Qi
rn
Oi
GO
CO
UJ
18
17
16
15
14
16
15
14
13
15
14
13
12
15
14
13
12
11
13
12
11
10
12
11
10
11
10
9
- QL = 401 £/min
200
QT = 352 £/min
QT = 310 £/min
_ QT = 273 £/min
QT = 227 £/min
QT = 180 £/min
= 136 &/min
H
3. 8 cm
23 cm
300
400
ur, cm/s
b
500
Figure 26.
Experimental pressure drop of a
3-stage mobile bed with hardware
screen support.
83
-------
UG - 500 cm/s) and can be neglected. The measured pressure drop
across the mobile bed is almost independent of gas flow rate when
the liquid rate is low. At higher liquid rate, the pressure drop
is slightly dependent on gas flowrates. These findings are consis-
tent with Wozniak (1975), Blyakher, et al., (1967), Barile and
Meyer (1973), and Epstein, et al.,'s results. W6zniak has found
that the pressure drops increase proportionally to the linear gas
velocity to the 0.2 power.
Figure 27 is a plot of the pressure drop versus liquid-to-
gas ratio with superficial gas velocity as parameter. The mobile
bed is 23 cm deep of 3.8 cm diameter spheres. The bed support
and retaining grids were hardward screens. As can be seen, the
pressure drop varies linearly with liquid-to-gas ratio. The
slope of the straight lines depends on the gas velocity.
Figure 28 shows the effects of static bed depth and number
of stages on pressure drop. The pressure drop is approximately
proportional to the number of mobile bed stages but not to the
static bed depth. The pressure drops increase proportionally to the
static bed depth to the 0.84 power.
Figure 29 shows the effect of packing sphere diameter on
pressure drop. As can be seen, pressure drop is higher for a
smaller ball diameter. This is expected since smaller packing
not only has a higher packing density, it also has higher liquid
holdup. Both of these contribute to higher pressure drop.
Plastic Net - Figure 30 shows the pressure drop data of the
plastic net supported mobile bed. Under the same operating condi-
tions, the plastic net support has a higher pressure drop than the
hardware screen support. The increase in pressure drop is not solely
due to the high friction loss of the plastic net support.
Two possible factors cause the pressure drop to rise with
the plastic net support. The first one is that liquid holdup in the
bed may be higher than with other support. In Chen and Douglas1
experiment, they only used one type of supporting grid, parallel
rods. Therefore, in their correlation the effects of the sup-
porting grid geometry on liquid holdup was not accounted for.
84
-------
u
E
U
CO
Of.
U3
(X
CL
O
o;
Q
w
Di
00
CO
tq
10
9
7
6
5
4
3
2
1
0
u.
500 cm/s
400 cm/s
340 cm/s
cm/s
PACKING DIAMETER = 3.8 cm
STATIC BED DEPTH = 23 cm
HARDWARE SCREEN SUPPORT
I I _J I I I I
/-v Up = 280 cm/s
/v u,, = 250 cm/s
u = 220 cm/s
8
10
11 12 13 14 15
Q /Qr x 103, cm3/cm:
Figure 27. The variation of pressure drop with liquid-
to-gas ratio and air velocity.
-------
12.5
10.0
7.5
PJ
PC;
CL
2 .5
150
-o—o-
QL = 136 a/min
d, = 3.8 cm diameter
b
HARDWARE SCREEN SUPPORT
2-STAGE MOBILE BED
HS = 30 cm
3-STAGE MOBILE BED
-Q H - 15 cm
2-STAGE MOBILE BED
H = 15 cm
s
I
I
200 250 300 350
SUPERFICIAL GAS VELOCITY, cm/s
400
Figure 28. Effects of bed depth and number of stages
on pressure drop.
86
-------
oo
CO
Pi
d.
o
Pi
w
pi
^2
CO
CO
w
Pi
10
9
8
7
6
5
4
3
2
I
T^
- HARDWARE SCREEN SUPPORT
HS = 23 cm
QT = 276 £/min
--o
0 1 ' ' i i i i i i i I i i i i i i i
I i i i I i i i i i I i
200
250
300 350 400 450
SUPERFICIAL GAS VELOCITY, cm/s
500
550
Figure 29. Effect of packing diameter on pressure drop
-------
34
30
3-STAGE MOBILE BED
H = 23 cm
d, = 3.8 cm
D
PLASTIC NET SUPPORT
20
u
o
pi
Q
CO
CO
PJ
10
I I
I I i I I I
- -O
200
Figure 30.
314 £/min-
261 Jl/min
196 £/min -
= 0 £/min
J ] I I I I i I L I I
300 400 450
ur, cm/s
Pressure drop vs. superficial gas velo-
city for 3-stage mobile bed with plastic
net support.
88
-------
Kito, et al. (1976c) used three different grids (8 mesh, 5 mesh,
and metal screen with 1.2 cm openings and 84% opening ratio) in
their experiments. They found that the geometry of the supports
has great effect on the amount of liquid holdup. Grids with
smaller openings give higher liquid holdup which in turn causes
the pressure drop to rise.
A second possibility is that liquid froth may be retained
on the plastic net. Russian researchers (Blyakher et al., 1967)
have observed that a liquid froth was retained on the grids which
had small openings. It is possible that grids with small openings
and small fractional open area act as sieve plates.
Figure 31 is the pressure drop across the three stage plas-
tic net supports; i.e., three stage mobile bed with packings
removed. This pressure drop is due to liquid retention on
the grids.
Figure 32 is a plot of the difference in pressure drop with
and without packing for the plastic net supported mobile bed.
The difference is equal to the pressure drops due to the weights
of packings and liquid holdup. As can be seen from Figure 32,
the pressure drop due to the weight of packing and liquid holdup
is independent of gas flow rate up to the flooding velocity. The
calculated AP, + APj, for the plastic net support is lower than
that for the hardware screen support. It is possible that Figure
31 is not a true representation of the liquid head on the grid.
When there are packings, some of the liquid froth will enter the
fluidized bed. Therefore, liquid holdup in the bed increases
and the retention on the grid decreases by the same amount.
Particle Collection -
Particle collection data have been reduced into plots of
particle penetration versus particle diameter for all experimen-
tal runs. Figures 33 through 35 are examples of the results.
As revealed by comparing the grade penetration curves, the
scrubber performance depends only on the overall pressure drop
across the scrubber. The grade penetration curves of two mobile
bed scrubbers, which have different bed geometries and operating
conditions, would be the same if they have the same overall pressure
89
-------
15
THREE STAGE MOBILE BED
WITHOUT PACKING
o
e
u
O
CO
LO
w
PS
Cu
10
QT - 363
314
196
200
Figure 31.
I
250 300 350 400
SUPERFICIAL GAS VELOCITY, cm/s
Pressure drop across four plastic net
supports.
450
90
-------
20
u
o
PS
to
oi
I — J
CO
en
tu
Pi
a,
15
10
0, = 363
-------
o
•H
U
CO
2
O
W
tu
1.0
0.5
0.1
0.05
0. 01
RUN NO. Cl-6
Cl-7
Cl-
Cl-9
SINGLE STAGE MOBILE BED
ur = 290 cra/s
b
QT/Q_ = 5.6 £/m3
L b
AP = 5.1 cm W.C.
d, = 3.8 cm
b
H = 30 cm
s
TITANIUM OXIDE AEROSOL
i I
1 5 10
AERODYNAMIC PARTICLE DIAMETER, ymA
50
Figure 33. Experimental grade penetration
curves.
92
-------
1.0
G
o
•H
•P
u
03
Si
M-i
2:
o
Ct,
t-
w
2:
m
OH
w
_i
u
0. 5
0.1
0.05
0. 01
2-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
ur = 260 cm/s
b
QL/QG =4.4 £/m3
AP = 5.0 cm W.C.
db = 3.8 cm
H = 15 cm
s
RED IRON OXIDE AEROSOL
i l i i i i i i i
1 5 10 50
AERODYNAMIC PARTICLE DIAMETER, ymA
Figure 34.
Experimental grade penetration
curves.
93
-------
1. 0
4-1
U
03
O
I— I
E-
tU
_3
CJ
I — I
H
Di
0.5
0.1
0.05
RUN NO. C3-61
3-STAGE MOBILE BED
u,, = 390 cm/s
Q /Qf = 7.5 £/m3
H = 23 cm
d, - 3.8 cm
b
AP = 15.4 cm W.C
CUPOLA DUST AEROSOL
I I I I I
0.5
i i r
RUN NO. C3-62
I
1 5
AERODYNAMIC PARTICLE DIAMETER, ymA
10
Figure 35. Experimental grade penetration
curves.
94
-------
drop. This characteristic is in line with mass transfer in mobile
bed scrubbers. After analyzing the S02 absorption data reported by
EPA (Borgwardt, 1972, 1974a, 1974b, and 1975) and Epstein (1976),
McMichael et al. (1976) and Wen and Chang (1978) concluded that
the efficiency for absorption of S02 in lime and limestone slurry
depends only on the pressure drop across the mobile bed scrubber.
Since the scrubber performance depends only on the scrubber
pressure drop, any parameter that leads to a higher pressure drop
will increase the collection efficiency accordingly. Pressure drop
may be increased by using higher liquid flow rate, heavier packings,
deeper bed, and supports with smaller openings or open area.
The particle collection of a 3-stage mobile bed scrubber cannot
be projected from the collection of a 1-stage scrubber; i.e.
Pts ^ Pti3. The projected penetration is slightly higher than
that actually measured (Figure 36). The cause may be due to particle
bypassing due to channeling. In a multiple stage scrubber, the
stages are attached one on top of another. This arrangement pro-
vides both mixing and a damping action which reduce particle bypassing.
Some inter-stage samplings were conducted on the multi-stage
scrubber. Results indicated that even though the particle collection
of a multi-stage scrubber cannot be projected from the collection
of a 1-stage scrubber, it can be projected from the collection of
one of its stages (Figure 37).
Particle collection in a mobile bed scrubber stems from inertial
impaction on the atomized liquid drops and on the packing elements.
For scrubbers which rely on inertial impaction for particle collec-
tion, the performance depends on the gas-phase pressure drop or
the power input to the scrubber.
Different mobile bed scrubber geometries and operating condi-
tions result in different scrubber performances and pressure drops.
It is very difficult to evaluate and compare the scrubber capability
based only on the grade penetration curves. A mobile bed scrubber
with plastic net supports has a higher collection efficiency than
a hardware screen supported mobile bed with the same bed geometry
and operated under the same parameters. It is not true that plas-
tic net is better than hardware screen since the pressure drop is
higher with plastic net.
95
-------
-M
U
2:
c
i—i
H
H
W
Z
ra
(X
w
1.0
0. 5
0.1
0.05
0.01
1-STAGE MOBILE
BED
AP = 4.9 cm W.C. .
w
3-STAGE MOBILE
BED
AP = 14.4 cm W.C.
w
UG = 240 cm/s
QL/QG = 9.2 £/m3
' 3-STAGE CALCULATED
* FROM 1-STAGE
\
\
db = 3.8 cm
HS = 30 cm
RED IRON OXIDE AEROSOL
. 1 . . . .1
1 5 10 50
AERODYNAMIC PARTICLE DIAMETER, pmA
Figure 36. 1-stage and 3-stage mobile bed
scrubber performance.
96
-------
o
•H
4->
O
ri
H
PJ
21
PJ
PL,
H
Ctf
<
PH
1.0
0. 5
0.1
0.05
0.01
STAGE 1 AND 2
3-STAGES
HARDWARE SCREEN SUPPORT
Up = 230 cm/s
QL/QG =11.2 Jl/m3
AP = 13.1 cm W.C.
w
d, = 3.8 cm
D
H - 23 cm
. . I . . ..I
J L
1 5 10 50
AERODYNAMIC PARTICLE DIAMETER, ymA
Figure 37. Penetration curves from interstage
sampling.
97
-------
A useful correlation called the cut/power relationship has
been developed by Calvert (1974) for scrubber evaluation purposes.
Calvert has shown that the scrubber performance cut diameter is
a very convenient parameter for describing the performance and
capability of a particle scrubber. Cut diameter is the particle
diameter whose collection efficiency is 501 and it can be obtained
from the experimental grade penetration curve. The cut/power
relationship is a plot of the cut diameter given by the scrubber
against pressure drop or power input.
Figure 38 is a plot of the measured performance cut diameter
of the mobile bed scrubber versus measured pressure drop for
various scrubber configurations and operating conditions. The
data fall on a straight line on log-log paper. The equation
describing the straight line is
dpc = 60.2 (APw) '-23 (29)
where d = scrubber performance cut diameter, ymA
AP = pressure drop, cm W.C.
The equation was obtained by linear regression and the
correlation coefficient is -0.94.
Comparison of data with other researchers' data - Figure
39 compares the published mobile bed performance data with that
of present study. As can be seen, data reported by Epstein et
al. (1975), Johnson et al. (1976), and Rhudy and Head (1977)
agree with that of present study. Data of Ensor et al. (1976)
show slightly higher efficiency.
Ensor et al. (1976) conducted a performance test on a
mobile bed scrubber which was installed in a coal burning power
plant. The scrubber was installed to supplement the particu-
late control by the electrostatic precipitator. The flue gas
from the precipitator (17,000 Am3/min at 137°C, 610,000 ACFM
at 280°F) enters a presaturator to reduce the gas temperature
to approximately 52°C (125°F). From the presaturator, the gas
enters the scrubber. At the scrubber outlet, the gas first
98
-------
10
<
w
u
u
>H
Q
O
1.0
0.5
1 I I i
HARDWARE SCREEN
SUPPORTED MOBILE BED
O
O
V
1-STAGE, d, =
2-STAGE, db =
2-STAGE, d, =
b
Q 3-STAGE, db =
3-STAGE, db =
3-STAGE, d, =
3.8 cm
3.8 cm
2.5 cm
3.8 cm
3.8 cm
5.1 cm
PLASTIC NET SUPPORTED
MOBILE BED
Q 3-STAGE,
= 3.8 cm
I
O
V
A
i i i i
I r
Figure 38.
5 10
PRESSURE DROP, cm W.C.
Experimental cut/power relationship for mobile bed scrubber.
50
-------
10
u
0.5
0. 1
A.P.T. PRESENT STUDY
A EPSTEIN ET AL.
OENSOR ET AL.
DJOHNSON ET AL.
ORHUDY AND HEAD
i i i i i i i i i
D
5 10
AP, cm W.C
50
100
Figure 39. Comparison between published mobile bed
performance data with present study.
100
-------
passes through a chevron-type mist eliminator then is heated by
steam coil to 85°C (185°F) before entering the stack. Particle
samples were taken before the presaturator and after the reheater.
Thus, Ensor et al.'s particle sample might not be a representa-
tive sample of what actually existed in the scrubber especially
since there was a presaturator in between. Particle growth
might occur in the presaturator.
Bed Expansion -
When the gas velocity in a mobile bed scrubber is maintained
above the minimum fluidization velocity, the bed expands. There-
fore, the retaining grids should be far enough apart to allow
this expansion. The operating bed heights were measured in this
study by observation. Since the operating bed height fluctuates
during a run, an average based on the judgement of the observer
was recorded. The data obtained by this method might not be accurate
However, they provide information on the general trends.
Figures 40 and 41 show the bed expansion as a function of the
liquid-to-gas ratio with gas velocity as a parameter for the 15 and
23 cm static bed depths; respectively. Figure 42 shows the effect
of packing size on bed expansion. Bed expansion is defined as:
H,-H
Bed expansion (%} - -|—- x 100% (30)
s
where H, = dynamic or operating bed height, cm
H = static bed height, cm
The results show that the bed expands linearly with increasing
liquid-to-gas ratio. The rate of expansion depends on gas flow
rate. The rate increases with increasing superficial gas velocity.
Under the same operating conditions, the 23 cm deep bed expands
less than the 15 cm deep bed. 5.1 cm diameter packings result in
a smaller expansion than the 3.8 cm diameter packing. These
phenomena are expected since the weight of the bed is higher with
the deeper bed and smaller packings.
Minimum Fluidization Velocity -
Chen and Douglas (1968) defined the minimum fluidization
velocity of a mobile bed as the maximum gas flowrate that will
101
-------
230
O
t— i
C/3
x
tu
<=>
w
PQ
200
150
100
50
—r-
O
A
T
u^, =
220 cm/s
250 cm/s
T 1 1 1 r
0 ur = 310 cm/s
ur = 340 cm/s
u
O UG
= 280 cm/s
V
O
HARDWARE SCREEN SUPPORT
H = 15 cm
PACKING DIAMETER = 3.8 cm
i
567 8 9 10 11 12 13 14
,3
QL/QG x 103, cm3/cm:
Figure 40. The variation of bed expansion with
liquid-to-gas ratio.
102
-------
230
2
O
HH
C/3
z;
x
w
Q
w
CO
200
150
100
50
= 500 cin/s
Q UG = 400 cm/s
u,, = 340 cm/s
© UG = 310 cm/s
/T\ UG = 280 cm/s
u = 250 cm/s
U = 220 cm/s
PACKING DIAMETER =3. 8 cm
H = 23 cm
s
HARDWARE SCREEN SUPPORT
0
10
15
QT/Qp x 103, cm3/cm:
JLr (j
Figure 41.
The variation of bed expansion with liquid-
to-gas ratio.
103
-------
13
o
I—I
H
i
O
Q
I-H
c/
h—(
t-J
11
80
I I I
STATIC BED DEPTH = 23 cm
2 . 5 cm dia . , u
5.1 cm dia. ,
©2.5 cm dia., u,
O 5.1 cm dia., u.
G
'G
450 cm/s
450 cm/s
310 cm/s
310 cm/s
\
I
100 120 140 160 180 200 220 240
BED EXPANSION, %
260
Figure 42. Effect of packing diameter on bed expansion.
-------
maintain the static packed bed height. They found that the
minimum fluidization velocity increases with increasing packing
diameter and decreases with increasing liquid rate. Their findings
are confirmed in this study as illustrated in Figures 43 and 44.
The minimum fluidization velocity was obtained by extrapolating
the bed expansion curves in Figures 40 and 41 to zero bed expan-
sion .
Slurry Scrubbing
Pressure Drop -
The overall pressure drop of a 3-stage mobile bed scrubber
with a limestone slurry as the scrubbing liquid are plotted in
Figure 45 along with cold run data. The addition of CaC03 to the
scrubber liquid has no effect on the scrubber pressure drop.
Particle Collection -
Runs #S3-1 through S3-9 of slurry scrubbing were planned
to investigate the effect of the slurry characteristics on the
scrubber particle collection mechanisms. The scrubber perfor-
mance cut diameters are listed in Table B-10. Figure 46 is a
plot of the scrubber performance cut diameter versus pressure
drop along with the cut/power relationship determined in the
cold operation mode. Since the data for the slurry tests fall
on the line of the cold run cut/power, it appears that the cal-
cium carbonate particles suspended in the scrubber liquid do
not affect the particle collection mechanisms and efficiency
of the mobile bed scrubber.
Runs #33-10 through S3-17 were intended to determine the
contribution of entrainment to particle emission. For these
runs, no aerosol was fed into the scrubber system. The particle
loading was determined with cascade impactors in the scrubber
outlet duct. As can be seen from Table B-10, the particle
loading due to entrainment is very low. The tube bank entrain-
ment separator in the mobile bed performed satisfactorily.
105
-------
(/)
E
o
o
J
tq
CO
i—i
Q
100
50
10
PACKING DIAMETER = 3.8 cm
STATIC BED DEPTH = 23 cm
I I I
100
200 100 400
LIQUID FLOW RATE, 5,/min
500
Figure 43. The variation of minimum fluidi-
zation velocity with liquid flow
rate.
106
-------
E
o
w
w
s
<
I—I
Q
I i I
/7\
= 363
= 227 £/min
I I
j i
50 100
MINIMUM FLUIDIZATION VELOCITY, cm/s
i I
150
170
Figure 44. The variation of minimum fluidization velocity with
packing sphere diameter.
-------
20
u
ex
o
ra
«
a,
15
10
3-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
d, = 3.8 cm
b
HS = 23 cm
SCRUBBER LIQUID: 5% § 10% LIMESTONE
SLURRY
COLD RUN DATA
o
Q =
QL = 227 £/min
200 300 400
SUPERFICIAL GAS VELOCITY, cm/s
Figure 45. Effects of slurry on pressure
drop.
108
-------
10
<
1.0
u
CX
DATA FOR COLD
OPERATION MODE
O DATA FOR SLURRY SCRUBBING
0.1
i i i i i i i
10
AP, cm W.C.
Figure 46. Slurry scrubbing test data.
100
109
-------
F/C Scrubbing
Pressure Drop -
The pressure drop data for F/C scrubbing are plotted in
Figure 47 along with data from cold operating mode. As can be
seen, F/C scrubbing will not increase the scrubber pressure drop.
Particle Collection -
The F/C runs were designed to study the feasibility of
using the mobile bed scrubber as a flux force/condensation
scrubber. A total of 27 runs were made with a three stage
scrubber. Each stage was packed with 3.8 cm diameter balls to a
depth of 23 cm. Twenty-two of the runs were using the hardware
screen support; the remaining five runs used the plastic net
support. The scrubber water flow rate was maintained at 273 Jl/min.
The overall scrubber pressure drop ranged from 12.6 cm W.C. to 15.0
cm W.C. The average pressure drop for all runs was about 13 cm W.C,
The experimental conditions are listed in Table B-ll and B-12.
The experimental grade penetration curves are presented in Appen-
dix E.
Figure 48 is a plot of performance cut diameter versus conden-
sation ratio. The condensation ratio is defined as grams of
water vapor condensed per gram of dry air in the scrubber. It
describes the maximum amount of water vapor per gram of air that
can be contributed to particle growth.
As can be seen from the figure, the cut diameter decreases
as the condensation ratio increases. The performance cut diameter
is reduced from 2.7 ymA with no F/C effect to about 0.7 ymA for
a condensation ratio of 0.25 g/g dry air, under approximately
the same scrubber operating pressure drop. This is a significant
improvement in performance per unit of power. To achieve a cut
diameter of 0.7 ymA, the required pressure drop for a mobile bed
scrubber without F/C effect is about 37 cm W.C. instead of 13 cm
W.C. with F/C.
Figure 49 shows the effect of condensation ratio on the pene-
tration of 1.0 ymA particles. The penetration decreased from
110
-------
e
u
fX
o
tq
oi
i — i
CO
C/3
PL,
15
10
COLD RUN DATA
G
3-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
db = 3.8 cm
H = 23 cm
QL = 273 £/min
q' = 0-0.25 g/g
200
300
400
SUPERFICIAL GAS VELOCITY, cm/s
Figure 47. F/C scrubbing pressure drop.
111
-------
IN)
Q
E-i
U
0.4
- 3-STAGE MOBILE BED
H = 23 cm
d, = 3.8 cm
D
AP = 13 cm W.C.
CUPOLA DUST AEROSOL
n = 1-5 x 107/cm3
l
I
I
I l I l
HARDWARE SCREEN
SUPPORT
PLASTIC NET
SUPPORT
0.01
0.05 0.1
q', g vapor condensing/g dry gas
0.5
Figure 48. The variation of cut diameter with condensation
rat io.
-------
o
•H
O
Pi
H
W
2:
w
PH
-H
u
1.0
0.5
0.1
5"
0. 01
HARDWARE SCREEN SUPPORT
PLASTIC NET SUPPORT
3-STAGE MOBILE BED
H = 23 cm
s
d, = 3.8 cm
b
AP = 13 cm W.C.
CUPOLA DUST AEROSOL, n = 1-5 x 107/cm3
J I
I I I
I I
d =1.0 ymA
0.05 0.1
q1, g vapor condensing/g dry air
0.5
Figure 49, Penetration for 1.0 umA diameter particle versus
condensation ratio.
-------
about 98% to about 401 when the condensation ratio increased
from 0 to 0.25 g/g dry air.
Particle Number Concentration Measurements -
The improvement in collection efficiency of a F/C scrubber
is due to particle growth and due to the addition of other
particle collection mechanisms, diffusiophoresis and thermopho-
resis. Diffusiophoretic deposition is caused by the condensa-
tion of water vapor from the gas onto a cold liquid surface
which exerts a force "sweeping" particles to the surface. Calvert
et al. (1973) have shown that the condensation ratio is sufficient
to define the particle deposition rate if there is no condensa-
tion on the particles. Whitmore (1976) concluded that the fraction
of particles removed from the gas by diffusiophoresis is equal to
either the mass fraction or the mole fraction condensing, depen-
ding on what theory is used for deposition velocity.
"Particle growth" means the enlargement of particle mass by
condensing a water film around the particle. The enlarged
particle is more susceptible to collection by inertial im-
paction. Particle growth is dependent on how well the particles
can compete with the cold surface for the condensing water and
the particle number concentration.
The particle number concentration was measured by using a
batch dilution system and two counters - a Gardner Condensation
Nuclei Counter (C.N.C.) and a Thermal Systems Electrical Mobility
Analyzer (E.M.A,) in the present study. The results of the count
experiments are shown in Table 8. The average inlet concen-
tration was found to be 5 x 107/cm3 using the C.N.C. The E.M.A.
data were obtained using all but the lowest size channel due to
electrical instability with this channel. In all but Run 1, the
E.M.A. count was lower than that of the C.N.C. in this test. The
outlet particle concentration was found to be 1.9 x 107/cm3 with
the C.N.C. for this run. The particulate loadings for these
series of runs are also given in Table 8.
The aerosol used for these experiments was redispersed cu-
pola dust which had a mass mean particle diameter of 1.6 umA.
114
-------
TABLE 8. PARTICLE COUNT RESULTS
Run Location C.N.C. E.M.A.
0
1
2
3
4
5
6
Ambient
Inlet
Inlet
Inlet
Inlet
Outlet
Outlet
2.
2.
7.
4.
5.
1.
2.
7x10
6x10
6x10
5x10
2x10
8x10
0x10
5
7
7
7
7
7
7
8.
3.
1.
2.
3.
4.
1.
7x1
3x1
1x1
1x1
2x1
2x1
0
0
0
0
0
0
4x10
i*
7
7
7
7
6
7
Notes
1. E.M.A. data obtained with channels 2-10
2. Inlet loading during runs 225 mg/DNm3
Outlet loading 110 mg/DNm3
115
-------
In contrast to the particulate existing at the foundry, the
redispersed dust is reasonably log normal and has a larger size
distribution. The large amount of submicron condensation type
aerosol is absent when redispersed.
The loading of the redispersed aerosol has varied signi-
ficantly during the F/C experiments from about 30 mg/DNm3 to
250 mg/DNm3 . The size distribution does not change signifi-
cantly with loading hence the assumption was made that the
particulate number concentration varies in direct proportion
to particulate loading.
Grown Particle Size Measurement -
The grown particle size is a function of particle number
concentration and the effective condensation ratio. The ef-
fective condensation ratio is defined as the total condensation
ratio, q1, multiplied by a factor, £ which indicates the
fraction of the water vapor which condenses on the particles.
Calvert and Gandhi (1977) have predicted that the fraction of
condensing vapor condensed on the particle could range from
about 0.1 to 0.4 for sieve plates and each individual particle
obtained approximately the same amount of condensate.
If the particle number concentration is known, the grown
particle size can be calculated by assuming that each individual
particle obtains the same amount of condensing water.
-
- - + -i (d x 10 ) > x 10 (31a)
772 n - P Pl
f q' + 3.78 x 10"10 n d * p \
where p = -2 2—PJ P— (31b)
P2 £p q' + 3.78 x 10"10 np d^p^
116
-------
d = initial particle diameter, ym
d = final particle diameter, ym
Pa
n = particle number concentration, #/cm3
£ = fraction of water vapor condensed on the
particles, g water/g gas
q1 = condensation ratio, g water/g gas
p = initial particle density, g/cm3
p = density of grown particle, g/cm3
Pz
There is no published information on the effective conden-
sation ratio in a mobile bed scrubber. Several particle growth
experiments were conducted to determine the extent of particle
growth and "f " for mobile bed scrubbers. This was accomplished
by measuring the particle size distribution simultaneously at
the scrubber inlet and at the top of the scrubber second stage.
A University of Washington cascade impactor was used for inlet
size distribution measurement. A special device designed by
A.P.T. was used at the top of the second stage.
The device consisted of a one-stage impactor followed by
a total filter and a regular sampling train for flow measure-
ment and control. The one-stage impactor was a modified
Greenburg-Smith impinger with a 0.16 cm diameter nozzle at-
tached to the inner glass tube. The impinger was filled with
100 ml of distilled water.
The sampling probe was a side port probe. To prevent
entrained water from entering the sampling device, an automatic
drainer was built into the probe. The experimental setup is
shown in Figure 50.
For each run, amounts of particles caught by the filter
and the one-stage impactor were determined. The fraction of
the total caught on the filter is the fraction of particles
which has diameters less than the cut-diameter of the one-stage
impactor.
117
-------
SIDE PORT
SAMPLING PROBE
oo
AUTO-
MATIC
DRAINER
LID
2ND STAGE DUCT
°
o
IMPINGER
COLLECTOR
FILTER
IMPINGER SET
SILICA GEL
SAMPLING TRAIN
Figure 50. The grown particle experimental set-up.
-------
Three runs were made, and the results are shown in Table
9 and plotted in Figure 51. The solid line in Figure 51 is
the predicted dry particle size distribution at the top of
the second stage. It was predicted from the measured inlet
particle size distribution and the combined grade penetration
curve of stages 1 and 2.
By assuming a "f ", the growth particle size distribution
can be predicted from equation (31) and the dry particle size
distribution. The growth particle size distribution was pre-
dicted for several "f ?s". When compared with data, it was
found that the predicted size distribution for f =0.15 agreed
with data, as shown in Figure 51. Therefore, the fraction of
condensate which condenses on the particles appears to be 0.15
for a mobile bed scrubber operated in the range of these
experiments.
119
-------
to
o
TABLE 9. PARTICLE GROWTH DATA
Run
1
2
3
q'
.24
.22
.25
Impinger
Cut Point
d ,ymA
1.1
1.6
1.6
Impinger
Collection
mg
12.4
5.4
2.6
Filter
Collection
mg
4.1
2.8
2.0
\ Caught on
Filter*
25
34
43
* Fraction caught on filter has particle size less than the cut diameter of
modified impinger.
-------
0.7
0. 5
0. 3
n
P
0.15, q' = 0.2
2.5 x 107/cm3
L
T
£
n
0.15, q' = 0.2
5 x 107/cm3
2ND STAGE DRY SIZE
DISTRIBUTION
INLET
LOADING q'
0
195 0.24
128 0.22
3} 133 0.25
I I
1 2
10 20 30 40 50 60 70
wt, I < d
P
90 95 98 99
Figure 51. The predicted and measured particle grown size
distribution.
-------
SECTION 6
MATHEMATICAL MODELING
LITERATURE SEARCH
A bed of low density spheres fluidized by upward flowing
gas and irrigated by downward flowing liquid may be used as a
gas-liquid contacting operation. The spheres may typically be
hollow plastic spheres of from 1 to 5 cm diameter and of den-
sity considerably below that of water.
The operation may be used for gas absorption in general
and should be expected to have favorable heat and mass transfer
characteristics because of the turbulent motion of the spheres.
It has also been used for scrubbing of industrial gases contain-
ing solid particles.
Fundamental studies on the mechanics of mobile beds such
as hydrodynamics, pressure drop, liquid holdup, minimum fluidi-
zation velocity and axial liquid mixing have been conducted by
a number of investigators. The following is a brief review of
the literature.
Hydrodynamics
The mobile bed scrubber consists of several packed beds
stacked inside a vessel shell. It is in many ways similar to
conventional packed towers. For a packed bed, if the pressure
drop across the bed is less than the sum of the weights of liquid
holdup and packing, the bed will remain stationary., However, if
the pressure drop is greater than the weight of packing plus
liquid holdup, the bed will expand and fluidization occurs; i.e.,
it becomes a mobile bed.
122
-------
O'Neill et al. (1972) have studied the hydrodynamics of
mobile bed scrubbers. They indicated that the mobile bed scrubber
can be operated in either of two modes, namely fluidization with-
out flooding and fluidization due to incipient flooding. For
mass transfer applications, operation in the incipient flooding
mode is desirable since it results in a higher gas-liquid inter-
facial activity.
The different modes of operation of a mobile bed can be
inferred from the traditional loading flooding curve of a packed
bed. The flow characteristics of a packed tower are usually
presented graphically as a series of curves similar to those
shown in Figure 52. Here, the pressure drop in the tower is
expressed as a function of the gas flow, with liquid flow as para-
meter. The curves are traditionally drawn with two distinct
changes in slope, and the break points "B" and "C" are known as the
loading and the flooding points, respectively.
Fluidization which occurred at flowrates below the flooding
point of a packed bed which has the same geometry as the expanded
fluidized bed is termed fluidization without incipient flooding,
Fluidization which occurred after the flooding point is termed
fluidization due to incipient flooding.
For the operation mode of fluidization without incipient
flooding, the minimum fluidization velocity is expected to increase
with packing density. If the fluidization occurred below the load-
ing point, the liquid holdup at a fixed liquid flowrate is expected
to remain approximately constant for different gas rate. Slight
dependence of holdup on gas flow rates is expected if the fluidiza-
tion occurs between the loading point and the flooding point.
For fluidization due to incipient flooding, the liquid hold-
up will increase with increasing gas flow rate and with increasing
density of the packing material. The minimum fluidization velo-
city is independent of the density of the packing material.
The mode of operation depends largely on the density of the
packing material and to a lesser extent on packing size, liquid
123
-------
o
Pi
Q
W
Qi
CO
CO
tu
c;
PH
L = CONSTANT
A FLOOD
C FLOODING
POINT
B/ LOADING
POINT
L = 0
LOG (GAS MASS VELOCITY)
Figure 52.
Typical pressure drop - flow
characteristics in conven-
tional packed towers.
124
-------
flow rate and physical properties of the liquid. Figure 53 given
by O'Neill, et al. shows the regime of the two operational modes.
It was derived from the assumption of a constant pressure gradient
at flooding of a static packed bed (20.8 cm W.C./m of packing,
2.5 in. W.C./ft) and from the application of Chen and Douglas'
correlation for liquid holdup. This figure will be useful in
predicting the mode of operation of the mobile bed scrubber.
The mechanics of the mobile bed are extremely complex.
Little theoretical work on the predictions of minimum fluidi-
zation, liquid holdup, and bed expansion has been published.
However, a large number of papers on experimental measurements
on mobile beds have been published. Numerous empirical corre-
lations are presently available to predict pressure drop, liquid
holdup, minimum fluidization velocity, interfacial area, and heat
and mass transfer coefficients. In using these empirical corre-
lations, care should be given to the scrubber operation mode and
the ranges of the variables which the researchers had used in
their experiments. The following is a summary of the empirical
correlations.
Minimum Fluidization Velocity -
The minimum fluidization velocities of mobile beds have been
investigated by Chen and Douglas (1968), Balabekov, et al. (1969,
1971), and Kito, et al. (1976c).
Chen and Douglas defined the minimum fluidization velocity
of a mobile bed as the maximum gas flow rate that will maintain
the static bed height. They measured the dynamic bed height,
"H," for various gas and liquid flow rates. They then plotted
the ratio of dynamic bed height and static bed height versus gas
flow rate for a constant liquid flow rate. A straight line was
obtained. They extrapolated the line to the ratio equal to 1
to obtain the minimum fluidization velocity. The empirical corre-
lation given by them for predicting the minimum fluidization velo-
city is:
G - = 0.106 d, J-15 10°'36L (32)
mf b
125
-------
0.5
0.4
IX)
o
H
i — i
C/D
2;
w
PH
0.3
0.2
0.1
REGION OF FLUIDIZATION DUE TO
INCIPIENT FLOODING
REGION OF FLUIDIZATION WITHOUT
INCIPIENT FLOODING
I
_L
0.5 1.0 1-5 2.0
LIQUID MASS RATE, g/cm2-s
1.3 cm •
2, 5
3.0
Figure 53. Region of mobile bed operation mode.
-------
where:
G £ = minimum fluidization gas mass velocity, g/cm2-s
d, = ball diameter, cm
b '
L = Liquid mass velocity, g/cm2-s
The packings used by Chen and Douglas were 1.26 cm, 2.54
cm, and 3.8 cm diameter polystyrene spheres. The densities were
0.16 g/cm3, 0.17 g/cm3, and0.16g/cm3 for 1.26 cm, 2.54 cm, and 3.8
cm diameter spheres, respectively. The ranges of gas and liquid
mass velocities studied were:
0 £ G £ 0.38 g/cm2-s
0 <_ L <_ 1. 5 g/cm2-s
Therefore, Chen and Douglas operated their mobile bed in the
nonflooding mode.
Balabekov, et al. (1969,1971) determined the minimum
fluidization velocity from pressure losses just as in conven-
tional fluidized beds, and followed the effects of liquid flow
rate packing size, and packing density on this minimum fluidi-
zation velocity. Their correlation is:
,, 2
a p
r
G
-
= 8 f exp
g PL PL
0.25 / Pp \
-------
Kito, et al. (1977c) used the same approach as Balabekov,
et al. and included the effects of liquid holdup and the geometry
of the supporting grid on the minimum fluidization velocity. They
correlated the minimum fluidization velocity by using the fluidi-
zation velocity for dry spheres. The correlation is expressed as:
= 3.14 f —
5 \ D
uGmfd
u
= 0.78
u
Gmfd
, £.
u
D
e
f —
> 0.05
d
D
< 0.05
(34)
(35)
where:
u
Gmf
u
Gmfd
"L
d
Dc =
f =
u
minimum fluidization gas velocity, cm/s
minimum fluidization velocity of dry packing, cm/s
superficial liquid velocity, cm/s
equivalent diameter of the pore on the supporting
grid, cm
column diameter, cm
fractional opening area of the supporting grid, fraction
minimum fluidization velocity for a packed bed with-
Gmfd'
out liquid flow is that calculated from Wen and Yu's (1966)
correlation.
1.75
(1 ~£ _c) P
mf G
UGmfd
e 3 d
L mf b
+ 150
["-'.f)1
L emf8
yG uGmfd
V J
b
(36)
128
-------
where:
uGm£d = minimuin fluidization velocity for dry spheres, cm/s
s <- = voidage of bed at minimum fluidization velocity,
fraction
du = ball diameter, cm
b
p, = packing density, g/cm3
PG = gas density, g/cm3
\IG = gas viscosity, g/cm-s
g = gravitational acceleration, cm/s2
The packings used by Kito et al. (1976c) have the following
properties:
db, cm
pv, g/cm3
1.16
0.76
1.95
0.17
1.95
0.54
1.95
1.16
2.85
0.29
2.85
0.59
The characteristics of the supporting grids were:
Grid
Grid
Grid
1
2
3
0
0
0
f
s
.712
.705
.84
de'
0.
0.
1.
cm
22
39
2
The superficial liquid flow rates in their experiments ranged
from 0 to 3.5 cm/s.
Strumillo, et al. (1974) observed that there is a distinct
effect of the column diameter and liquid flow rate on the mini-
mum fluidization velocity. They presented an empirical equation
for the calculation of the minimum fluidization velocity. The
equation was based on limited data and a small column diameter.
Therefore, it is applicable only to their system.
Tichy and Douglas (1972) measured the expanded bed heights
in a mobile bed for polystyrene spherical packings of two sizes,
1.25 cm and 1.9 cm in diameter, packing densities from 0.153 to
129
-------
0.458 g/cm3, and fixed bed heights from 14 to 35 cm. The sup-
porting grid was 0.24 cm diameter rods spaced at 1.25 cm apart.
The free area of the grid was 78%. They derived an empirical
equation for predicting the bed height. The minimum fluidization
velocity can be obtained by letting Hd/Hs = 1; i.e., by using Chen
and Douglas1 (1967) definition of minimum fluidization velocity.
Gmf = 3.16 (0.115 + 18.33 db - 0.5852L06
where:
G ,- = minimum fluidization gas mass velocity, g/cm2-s
m±
d, = ball diameter, cm
L = liquid mass velocity, g/cm2-s
Blyakher, et al. (1967) gave the following equation for
predicting the minimum fluidization velocity.
6.7 x 10"3ur .p, UT °'9
„ . Gmfd L f,R>
uGmf ~ uGmfd 1 . n ,7 „ 09 l J
1 + 0.67 Uj •
where:
Upinf = minimum fluidization velocity, cm/s
Up ,., = minimum fluidization velocity of dry packing, cm/s
UT = liquid velocity, cm/s
Lt
The minimum fluidization velocity of dry packing in equation
(38) was derived from Ergun's equation for packed bed pressure
drop. The gas velocity at which the pressure across the packed
bed equals the weight of the dry packing is "Up JTJ".
0.80 d,_°-715 f0l. -D^) °-572
u,
Gmfd 0.11*3 0429
where:
d, = ball diameter, cm
p, = packing density, g/cm3
130
-------
PG = gas density, g/cm3
yQ = gas viscosity, g/cm-s
The packings employed by Blyakher, et al. (1967) were 3.8
cm diameter hollow polyethylene spheres (packing density = 0.17
g/cm3) and 3.8 cm cellulose acetate spheres (packing density =
0.090g/cm3). Supporting grids with 19, 30, 41 and 90% open area
were studied. The grid which had an open area of 901 was a net-
work of steel rods with diameters of 0.2 cm and with a spacing
of 2.5 cm between them. The remaining grids had holes with
diameters varying from 1.4 to 2.4 cm.
Liquid Holdup -
Chen and Douglas (1968) determined the liquid holdup in a
mobile bed scrubber from the dynamic response curve for tracer
injection. They found that the liquid holdup related to the
fixed bed column is approximately independent of the gas flow
rate. It is a function of liquid flow rate and the packing
diameter. Their empirical correlation for liquid holdup in
a mobile bed is:
hLo = 9.44 x 10"2 db-°-5 L°'6 + 0.02 (40)
where:
hT = liquid holdup based on fixed bed volume, cm3/cm3
-LO
L = liquid mass velocity, g/cm2-s
d, = ball diameter, cm
The apparatus used by Chen and Douglas is the same as that
presented earlier. Therefore, this correlation is limited to
mobile beds which are operated in the nonflooding mode.
Kito et al. (1976d), using the same setup presented in
the last section, also found that the liquid holdup is indepen-
131
-------
dent of gas flow rate. In addition to liquid flow rate and packing
diameter, they found that the geometry of the retaining grid
also affects the amount of liquid holdup in the bed. Their
empirical equation is:
/d \ -°"84
hT = 0.06 + 0.0316 f -°-42 |-i] (dhr°-84 (p,)°-18(HJ-^ UT
LO S \D / DSL
C (41)
where:
hT = liquid holdup related to fixed bed volume, cm3/cm3
f = fractional open area of the supporting grid, fraction
d = equivalent diameter of grid opening, cm
D = column diameter, cm
d, = ball diameter, cm
p, = packing density, g/cm
H = static bed height, cm
UT = superficial liquid velocity, cm/s
LI
The liquid holdup predicted by Kito, et al.'s correlation
is about three times higher than that predicted by Chen and
Douglas1 correlation. The main reasons are probably that Kito
et al. used grids with smaller openings and packings with high
packing density. The supporting grid used by Chen and Douglas
was 0.24 cm diameter rods spaced at 1.25 cm apart. Therefore,
in Chen and Douglas' setup, no water was retained on the grid.
In Kito et al.'s setup, the supporting grids had much smaller
openings. As revealed in the experiments conducted in the pre-
sent study on the plastic net support, water was retained on
the grid even though there was no packing.
Bed Expansion -
Tichy and Douglas (1972), Balabekov, et al. (1972), Blyakher,
et al. (1967), and Kito, et al. (1976d) derived empirical equa-
tions to predict the expanded bed height of mobile bed scrubbers.
Their equations are:
132
-------
Tichy and Douglas (1972):
= 0.8849 + 3.166 G - 0.1833 d, + 0.233 L°'6 d °-5 (42)
D h ^ -'
Hd
H
Balabekov, et al. (1971):
(1-e) H + HT
1 - h,
H
Kito, et al. (1976d) :
Hs . 1 - hLQ
(43)
Blyakher, et al. (1967):
d = 1.17 + (6.5xlO-3 + 7.8x10-" u. °-" ) (un-u_) (44)
Hd 1 - 0.151 pG
where :
UQ = superficial gas velocity, cm/s
Hj = dynamic or expanded bed height, cm
H = static bed height, cm
o
G = gas mass velocity, g/cm2-s
L = liquid mass velocity, g/cm2-s
d, = ball diameter, cm
u, = superficial liquid velocity, cm/s
uGmf = minimuin fluidization velocity, cm/s
hLo = liquid holdup, cm3/cm3
e = void fraction of dry static bed, fraction
hG = gas holdup, cm3/cm3
HT = height of liquid column retained on the supporting
grid, cm
133
-------
The packing diameters, packing densities, and supporting
grid configurations used by these investigators were given in
earlier sections.
Gas Holdup -
Kito, et al. (1976a) investigated the effects of operating
parameters on gas holdup in mobile beds. They found that only
the gas flow rate and the liquid surface tension influenced the
holdup.
hG !«-• °-5NWe"' Npr"« (46)
f T
= Weber Number = —— (47)
N = Froude Number = ^-g- (48)
where:
hp = gas holdup, cm3/cm3
D = mobile bed column diameter, cm
u^ = superficial gas velocity, cm/s
PL = liquid density, g/cm3
aT = surface tension of liquid, dyne/cm
LJ
g = acceleration of gravity, cm/s2
The above equation can be approximated by:
°'lflf
= 0.055 u 'ff (49}
Flooding Velocity -
Balabekov, et al. (1971) presented the following equation
for the determination of flooding velocity of mobile beds:
(50)
1.16 e HL
134
-------
where:
u^r = flooding velocity, cm/s
UG' = original entrainment velocity in wetted packing, cm/s
H = packed column height or the distance between
retaining grids, cm
HL = height of clear liquid column retained on retaining
grid, cm
e = void fraction of dry packed bed, fraction
H = static bed height, cm
Uchida, et al. (1977) defined that the flooding point of
a mobile bed is reached when the terminal settling velocity of
the wetted packing is equal to the gas velocity. Since the dia-
meter of the packings in the mobile bed is large, the relation
between the flooding velocity and the minimum fluidization velo-
city can be approximated by the following equation:
urf
rp1 = 8.72 (51)
uGmf
where:
Upr = flooding velocity, cm/s
uGmf = minimum fluidization velocity, cm/s
Liquid-Gas Interfacial Area -
Wozniak and Ostergaard (1973) and Wozniak (1977) derived
an empirical equation for the calculation of liquid-gas inter-
facial area of mobile beds. The mobile bed scrubber was a two-
stage mobile bed, each packed with 1.96 cm diameter polypropylene
spheres (packing density 0.266 g/cm3). The supporting grids
were wire mesh with approximately 60% open area. The diameter
of the scrubber was 20 cm.
From the data of absorption of C02 by NaOH solution, they
back calculated the interfacial mass transfer area. The effec-
tive interfacial area, pressure drop, amount of gas and liquid
holdup are correlated by the following equation:
135
-------
0.8 0 2 2 0.9 3 3 7
hr \ /H, UP_.)\
J-= 9.2 x 10-" —5-1 M - (52)
V UG UG '
where:
a = effective interfacial area per unit volume of static
packing, cm2/cm3
a, = geometrical surface area of static bed per unit volume
of static packing, cm2/cm3
hG = gas holdup, cm3/cm3
H, = dynamic bed height, cm
AP = pressure drop across bed, cm W.C.
uf = superficial gas velocity, cm/s
\ir = viscosity of gas, g/cm-s
b
Kito, et al. (1976b) used the same technique to determine
the liquid-gas interfacial area. They found that the interfacial
area, on a tower volume basis, increases proportionately to
the gas flow rate up to 200 cm/s , and to gas holdup to 0. 6, and is
not affected by the geometry of the supporting grid, the static
bed height, and the packing density. The interfacial area
decreases with an increasing gas flow rate over 200 cm/s and gas
holdup over 0.6. No equation was given by them for the prediction
of interfacial area.
Pressure Drop
For a fully fluidized mobile bed, the pressure drop is made
up of the sum of pressure drops due to weight of dry packing,
weight of liquid holdup in the bed, the friction loss of the
retaining grids and column, liquid head retained on the supporting
grid, and drop atomization (Blyakher, et al., 1967; Kito, et al.,
1976) ; i.e.:
w - APb + APLh + APL + APa + APc + APf (53)
136
-------
where:
AP = overall pressure drop per stage of bed, cm IV. C.
AP, = pressure drop due to weight of dry packing, cm W.C.
APL, = pressure drop due to liquid holdup, cm W.C.
A?L = pressure drop due to liquid head retained on supporting
grid, cm W.C.
AP = pressure drop due to drop atomization, cm W.C.
cl
AP = pressure drop due to wall fraction, cm W.C.
AP = friction loss of supporting grid, cm W.C.
Pressure Drop Due to Weight of Dry Packing -
The pressure drop due to the weight of dry packing can be
expressed as:
APb = (Pb - PG)d-e] Hs (54)
where:
pi = packing density, g/cm3
PG = gas density, g/cm3
e = static bed porosity, fraction
H = static bed height, cm
Pressure Drop Due to Liquid Holdup -
The pressure drop due to the weight of liquid holdup in
the bed is:
APT, = pT hT H, = pT h H
Lh L L d L Lo s
where:
PL = liquid density, g/cm3
hL = liquid holdup, cm3/cm3
H, = dynamic bed height,cm
hLo = licluid holdup related to static bed, cm3/cm3
H = static bed height, cm
137
-------
Chen and Douglas (1968) and Kito, et al. (1976) determined
the liquid holdup in a mobile bed. Their empirical correlations
are given in an earlier section, Chen and Douglas (1968) and
Kito, et al. (1976) both found that the liquid holdup is inde-
pendent of gas velocity. Therefore, by using their correlation
in equation (55), the pressure drop due to liquid holdup should
be independent of the gas flow rate.
Uchida, et al. (1977) modified Kito, et al.'s liquid holdup
correlation to include the experimental data obtained from a
large scale mobile bed scrubber. Their correlation for pressure
drop due to liquid holdup is:
•SY"2I
'*e\-
*J
8 4
-0.64
db f
0.1 8
>b Hs
UL
^c'
where:
y, = viscosity of liquid, g/cm-s
f = fractional opening area of the supporting grid, fraction
d = equivalent diameter of grid opening, cm
D = column diameter, cm
c
d, = ball diameter, cm
p, = packing density, g/cm3
H = static bed height, cm
u, = superficial liquid velocity, cm/s
Wofniak (1977) performed a dimensional analysis and conducted
experiments to determine the coefficients. His correlation for
pressure drop due to liquid holdup is:
/H y»515/d „ ,1.798 /d 0T\°-8261
APLh - 476.6 PG ^ U JL^G. ^__L^
\db/ * yG / \ VL /
(57)
where:
APLh = Pressure drop due to liquid holdup, cm W.C.
PG = gas density, g/cm3
UG = superficial gas velocity, cm/s
138
-------
H = static bed height, cm
d, = ball diameter, cm
yp = gas viscosity, g/cm-s
yL = liquid viscosity, g/cm-s
UT = superficial liquid velocity, cm/s
J_j
PT = liquid density, g/cm3
Li
Blyakher, et al. (1967) and Balabekov, et al. (1971) proposed
other equations for calculating the pressure drop due to liquid
holdup. Their equations contain empirical coefficients which
need to be determined experimentally.
Pressure Drop Due to Liquid Froth Retained on Supporting Grid -
"APL", the pressure drop due to liquid froth retained on the
supporting grid is given by the following equation (Blyakher, et
al. 1967):
1.75 0.5
APL = ?f UG UL (58)
where:
UG = superficial gas velocity, cm/s
UL = superficial liquid velocity, cm/s
£,. = coefficient dependent upon the geometric characteris-
tics of the grid, dimensionless
Supporting Grid and Column Friction Loss -
The friction losses of the supporting grid and the column
can both be expressed by the following equation (Perry, 1973):
Pr urz
AP, or AP = f ——— (59)
° 2g
where:
f = hydraulic resistance coefficient, dimensionless
Pressure Drop Due to Drop Atomization -
The pressure drop due to drop atomization is (Calvert, 1968):
139
-------
(60)
where:
QT = liquid volumetric flowrate, cm3/s
Qf = volumetric gas flowrate, cm3/s
k = constant characterizing the fraction of liquid being
atomized, fraction
Heat and Mass Transfer Coefficient
The heat and mass transfer capabilities of mobile bed scrubbers
were studied by Douglas (1964) and Gel'perin, et al. (1973).
Douglas (1964) conducted two series of experiments. The first was
the absorption of ammonia by boric acid solutions. The results
were given in terms of height of transfer unit and as mass transfer
coefficients. Both of these performance values were calculated
using the overall gas logarithmic mean driving force and were based
on the static bed height. He found that the height of transfer
unit decreased with increasing liquid mass velocity and increased
with increasing gas mass velocity.
The second series of experiments were designed to obtain
the rate of simultaneous heat and mass transfer. The technique
used was the dehumidification and cooling of hot saturated gas.
He found that the effect of gas velocity on the height of trans-
fer unit was the same as that observed for the absorption of
ammonia. However, the effect of liquid mass velocity was dif-
ferent for the two cases. The height of transfer unit, instead
of decreasing with liquid rate, increased with liquid rate for
the dehumidification tests.
Douglas (1964) presented experimental data but no mathematical
correlations. Gel'perin et al. (1973) studied the process of
adiabatic evaporation of water during its contact with hot gas
in a mobile bed scrubber. The heat and mass transfer coefficients
referred to the unit volume of the expanded bed were given by
Gel'perin, et al. as follows:
140
-------
kr = 1.07 x ID'9 urL3 UT °-3 H °-4 (61)
(j b L S ^ J
hp = 3.9 x 10~13 Up1'3 UT °-3 H °* (62)
b b L S
where:
&
k~ = mass transfer coefficient,
cm3-s-atm
hG = heat transfer coefficient, kcal/cm3-s-°C
u,-, = superficial gas velocity, cm/s
b
UT = superficial liquid velocity, cm/s
Li
H = static bed height, cm
The ratio of heat transfer coefficient to mass transfer
coefficient is 3.65 x 10"1* kcal-atm/g-°C. According to
Gel'perin et al., the dynamic height of the mobile bed depends
on the same parameters as the coefficients of heat and mass
transfer and the mass transfer coefficient based on unit ex-
panded bed volume is independent on the liquid rate.
Particle Collection
The only model available for particle collection in a
mobile bed is the semi-empirical relationship presented by
Bechtel Corporation in a June 1971 report on the Shawnee project
for the EPA and was cited by Calvert et al. (1972). This cor-
relation is based on the premise that collection efficiency is
due to inertial impaction on the balls.
Pt, = exp - 9.5 x 10 - (u
(63)
141
-------
where:
Ptd = particle penetration for particles with diameter dpa,
fraction
Q = volumetric liquid flowrate, cm3/s
Q = volumetric gas flowrate, cm3/s
u = superficial gas velocity, cm/s
G
p = gas density, cm/s
n = number of mobile bed stages, dimensionless
H = static bed height, cm
d, = ball diameter, cm
b
K = inertial impaction parameter, dimensionless
d = aerodynamic particle diameter, ymA
pa
Up- = gas velocity in bed, cm/s
Vip = gas viscosity, g/cm-s
COMPARISON OF EXPERIMENTAL DATA WITH PREDICTIONS
Pressure Drop
Pressure drop predictions by correlations of Chen and
Douglas (1969], Wozniak (1977), Kito et al. (1976d), and
Uchida et al. (1977) were compared with the experimental data
obtained in this study in Figures 54 through 57. The experi-
mental data have been corrected for friction losses.
The comparison revealed that predictions by Uchida et al.'s
correlation are much higher than that measured. Predictions by
Kito et al.'s correlation are better than Uchida et al.'s cor-
relation; but it still is higher than that measured. Wozniak's
equation overestimated the pressure drops of the mobile bed with
hardware screen support and underestimated the pressure drops of
the mobile bed with plastic net support. The predictions by Chen
and Douglas agree only with the data for the hardware screen sup-
ported mobile bed. However, the correlation by Chen and^Douglas
did not predict the correct dependence of pressure drop on liquid
flow rate (Figure 58). Compared to the pressure drop data on
the plastic net supported mobile bed, the correlation by Chen
and Douglas is much lower than that measured.
142
-------
20
e
u
PH
O
Pi
Q
Pi
£D
CO
CO
w
Pi
a.
a
w
F-
u
I-H
Q
W
50
l i I I |
PLASTIC NET SUPPORT
V 3-STAGE,
cm,
V V
y
V VV
HARDWARE SCREEN SUPPORT
3-STAGE,
3-STAGE,
= 3.8 cm,
= 3.8 cm,
= 23 cm
= 15 cm
3-STAGE, d = 5.1 cm, H = 23 cm
O 2-STAGE, db = 2.5 cm, Hg = 23 cm
Zk 2-STAGE, dfa = 3.8 cm, Hg = 15 cm
A 1-STAGE, db = 3.8 cm, H = 30 cm
V .
10 20
MEASURED PRESSURE DROP, cm W.C.
30
40
Figure 54. Predicted and measured pressure drop (Chen and Douglas's correlation)
-------
o
d,
o
OS
oS
ED
CO
CO
w
Q
tq
Q
W
as
(X
I 1 1
vvv
HARDWARE SCREEN SUPPORT
Q 3-STAGE, d, = 3.8 cm, H = 15 cm
XV
3-STAGE, d, = 5.1 cm, H = 23 cm
__ D S
O 2-STAGE, db = 2.5 cm, H = 23 cm
2-STAGE, d = 3.8 cm, H = 15 cm
A 1-STAGE,
= 3.8 cm, H = 30 cm
PLASTIC NET SUPPORT
V 3-STAGE, d = 3.8 cm, H = 23
MEASURED PRESSURE DROP, cm W.C.
Figure 55. Predicted and measured pressure drop (Wozniak's correlation).
-------
20
& 15
u
O
Pi
O
tq
O)
w
Pi
U
pq
OS
CH
10
i i i
O 3-STAGE M.B., d, = 3.8 cm,
H =23 cm b
Q 3-STAGE M.B., d, =3.8cm,
HS = 15 cm b
> 3-STAGE M.B., d =5.1 cm,
H = 23 cm
i I I
O 2-STAGE M.B., d =2.5cm'
H = 23 cm
s
2-STAGE M.B., d = 3.8 cm.
H = 15 cm
s
A 1-STAGE M.B., d =3.8 cm-
H = 30 cm
s
I i
J I
I I
5 10 15
MEASURED PRESSURE DROP, cm W.C.
20
Figure 56. Predicted and measured pressure drop
(Kito et al.'s correlation).
145
-------
30
I I I
25
Q 3-STAGE M.B., db = 3.8 cm, Hg = 23 cm
3-STAGE M.B., d, = 3.8 cm, H = 15 cm O
3-STAGE M.B., db = 5.1 cm, HS = 23 cm
HARDWARE SCREEN SUPPORTED
MOBILE BED
O
O
Pi
O
01
G
PJ
LO
LO
tu
oi
fX
OS
u
I — I
Q
W
20
15
10
0
Q 2-STAGE M.B., d =2. 5 cm
H = 22 cm
2-STAGE M.B., d = 3.8cm
H = 15 cm
1-STAGE M.B., d, = 3.8 cm
H = 30 cm
' - 1 - 1 - 1 - 1
1 - 1 - 1 - 1 _ i ' » ' i i
Figure 57.
5 10 15
MEASURED PRESSURE DROP, cm W.C.
Predicted and measured pressure drop
(Uchida et al.'s correlation).
20
146
-------
e
U
a,
o
OS
Q
w
OS
!=)
CO
CO
Figure 58.
18
17
16
15
14
16
15
14
13
15
14
13
12
15
14
13
12
11
13
12
11
10
13
12
11
10
12
11
10
9
QL = 401 H/ruin
- QL = 31°
QT = 273 £/min
Li .
Q = 227 H/min
Li
_ QT =180 £/min
O
. Q = 352 £/min
PREDICTED
CHEN $ DOUGLAS
O
-O—O
Qr = 136 £/min
"" lj
^.
O
•o
200
300
400
ur, cm/s
b
500
Measured and predicted pressure drop for a three-
stage mobile bed scrubber with hardware screen
support,and each stage packed with 3.8 cm dia.
spheres to a depth of 23 cm.
147
-------
Particle Collection
The predicted particle collection in the mobile bed scrubber
by equation (63) did not agree with the experimental data ob-
tained in this study. The predictions were lower than that
measured.
Bechtel Corporation's correlation; i .e., equation (63) was
derived based on the premise that particle collection is due to
inertial impaction on balls. We may note that the impaction
parameter has a value of about 5 x 10 "^ for a gas velocity of
305 cm/s (10 ft/s), ball diameter of 3.8 cm, and aerodynamic
particle diameter of 1 . 0 ymA. The collection efficiency for
a sphere is 0% for values of the impaction parameter smaller
than about 0.1; consequently, it is impossible to attribute
collection efficiency to this mechanism.
Another drawback of equation (63) is that it did not in-
clude the effect of supporting grids on particle collection.
As mentioned earlier, the efficiency of a plastic net supported
mobile bed is higher than that of a hardware screen supported
mobile bed with the same bed geometry and liquid and gas flow rates .
MATHEMATICAL MODELING
Pressure Drop
An empirical correlation was derived in this study for the
prediction of pressure drop due to liquid holdup. The correla-
tion was derived by dimensional analysis. The correlation is:
(64)
where ?Lh = pressure drop due to liquid holdup in bed,cmW.C.
PG = gas density, g/cm3
PT = liquid density, g/cm3
pb = packing density, g/cm3
HS = static bed height, cm
dfe = packing diameter, cm
UG = superficial gas velocity, cm/s
148
-------
UT = superficial liquid velocity, cm/s
Li
yp = gas viscosity, g/cm-s
PT = liquid viscosity, g/cm-s
An empirical correlation was also derived in this study for
the prediction of pressure drop due to the liquid froth retained
on the grid.
APT - 9.2 x 10-9 —i- u "u" (65)
If the friction losses are neglected, the pressure drop
across one stage of the mobile bed is:
AP - (l-e)pb HS + APLh + APL (66)
For a multi-stage mobile bed scrubber, the overall pressure drop
will be:
AP = n AP + APr + AP (67)
w ± c
where AP = Overall pressure drop, cm W.C.
AP£ = friction loss of retaining grids,cm W.C.
AP = wall friction loss, cm W.C.
The predictions by the above equation are compared with data
reported by Epstein et al. (1974), Barile and Meyer (1971), Kito
et al. (1976), Blyakher et al. (1967), Tichy et al. (1973), and
Douglas et al. (1963) in Figures 59 and 60. The data reported by Epstein
et al. were obtained on a full scale mobile bed which was installed
in a power plant. They used two types of support, mesh and bar-
grid. The bars were 0.95 cm (3/8") diameter stainless steel and
spaced at 3.2 (1-1/4") on centers. The wire diameter of the
wire mesh grid was 0.38 cm (0.148"). The spacing between the wire
was not reported. In the pressure drop predictions, the wire
mesh was assumed to have the same geometry as the bar grid; i.e.,
same "f " and "d ". As can be seen from Figures 59 and 60, the
O G
agreement between predictions and data is good.
149
-------
25
o
OH
O
OS
Q
PJ
C/}
w
Qi
Q
W
pi
3
C/3
<
W
20
15
10
1 1
WOZNIAK
V EPSTEIN ET AL.
KITO ET AL.
O BLYAKER ET AL.
TICHY ET AL.
<0>BARILE AND MEYER
B DOUGLAS ET AL.
5 10 15 20
PREDICTED PRESSURE DROP, cm W.C.
25
Figure 59. Comparison between measured and predicted
mobile bed pressure drop.
150
-------
10
o
r,
OH
O
Pi
Q
W
Pi
S
CO
CO
WIRE MESH SUPPORT
H = 25 cm
s
d, = 3.8 cm
b
Pb = 0.174 g/cm3
~1.2 - 1.3 cm/s
0. 86 cm/s
PREDICTED
d = 2.2 cm
e
£ = 0.5
s
D = 195 cm
c
u =2.5 cm/s -I
7-2 cm/s
100
200
300
400
500
, cm/s
Figure 60.
Pressure drop through a single stage
of a mobile bed obtained at EPA/TVA
Shawnee plant.
151
-------
Particle Collection
Particle collection in a mobile bed scrubber may be due to
inertial impaction on the atomized liquid drops. Thus, particle
collection can be predicted if the atomized drop diameter and
amount of liquid in drop form can be calculated. However, the
hydrodynamics of a three-phase fluidized bed are extremely com-
plex. It is impossible to derive the theoretical equations for
drop size and quantity of drops. Empirical approximations based
on pressure drop relationships for gas atomized scrubbers were
used to predict collection efficiency without success. An em-
pirical equation was developed for particle collection in a
mobile bed scrubber. The empirical equation is:
Ptd = exp[-9.84 x KT^AP^1-96 d L6] (68)
where Ft-, = penetration for particle diameter "d", percent
or fraction
AP^ = overall pressure drop across the mobile bed
scrubber, cm W.C.
d = aerodynamic particle diameter, ymA
pa
F/C SCRUBBING
When a hot and saturated gas is in contact with cold water
or a cold solid surface, condensation of water vapor occurs.
Part of the vapor will be condensed on the particles which
serve as condensation nuclei. Thus, the particles will have
grown in mass due to the layer of water they carry and will be
more susceptible to collection by inertial impaction. While con-
densation occurs, there will be diffusiophoretic and thermo-
phoretic deposition on the cold surfaces as well as some inertial
impaction. The particle growth by condensation in combination
with diffusiophoresis and thermophoresis is referred to as
"flux force/condensation" (F/C) scrubbing.
152
-------
Several phenomena are simultaneously involved in a F/C
scrubber and the mathematical model is complex and cumbersome.
Calvert and Gandhi (1977), through a series of studies, con-
cluded that the flux force effects and condensation effect can
be treated separately. Based on this conclusion, they developed
a simplified performance prediction and design method. Their
method is summarized in the following paragraphs.
Basic Concepts
Before proceeding to the details of the mathematical model,
the basic concepts and outline of the approach will be discussed.
If we consider a typical F/C scrubbing system, it might have the
features shown in Figure 61. 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 surface is employed to cause condensation. While conden-
sation occurs there will be diffusiophoretic and thermophoretic
deposition as well as some inertial impaction (and, perhaps,
Brownian diffusion). The particles in the gas leaving the con-
denser 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
greater inertia of the particles, There may be additional con-
densation, 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. The condenser may be a separate
153
-------
CLEAN GAS
55°C
0-12 g/g
WATER —£>
IMPACTOR
55°C
0.12 g/g
WATER —{>
GAS
CONDENSER
SAT.A GAS
74°C
0. 36 g/g
WATER 0
SATURATOR
1
0
HOT Q GAS
,000°C
•01 g/g I
Figure 61. Generalized F/C scrubber svstem.
154
-------
unit or can be part of the scrubber as in the pilot plant of
the present study.
The efficiency of heat and mass transfer is high for
mobile bed scrubbers; most of the condensation occurs in the
first stage. In subsequent stages, the gas is scrubbed by
inertial impaction and there will be a minor amount of ad-
ditional condensation.
Diffusiophoretic Deposition
Particle deposition by diffusiophoresis was described by
the following equation (Calvert et al. 1973, 1975, 1976):
VM! DG / dy \
u n = r — — [ — , cm/s (69)
[y VMT + U-y) VMI] U-y) Ur /
pD
or,
/ 1 \ dy
UPD = Cl DG
,1-y/ dr
where D~ = diffusivity of water vapor in carrier gas, cm2/s
Mj = molecular weight of water, g/mol
M2 = molecular weight of nontransferring 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" described 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
fraction in air ranging from 0.1 to 0.5, "(V varies from 0.8
to 0.88. Calvert and Gandhi (1977) used a rough average of
0.85 for "Ci" for computing "u " and consequent particle col-
lection efficiency by integrating over the period of condensation,
155
-------
Whitmore (1976) 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 condensation
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 frac-
tion of the gas is transferred to the liquid phase it will carry
along its load of suspended particles.
Calvert and Gandhi (1977) have used Whitmore's general con-
cept but with two modifications. First, one can see from equation
(69) that Whitmore's theory would be comparable to assuming that
the particles move with the same velocity as the gas phase. Cal-
vert and Gandhi have chosen to retain the correction for molecular
weight gradient, which means that they will compute the particle
collection efficiency as 85% of the volume fraction of gas
condensing on the cold surface.
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. The fraction of the
condensate which causes particle growth depends on several
factors and ranged from about 0.1 to 0.4 of the total condensate
for the range of parameters they explored.
If one is concerned only with diffusiophoretic deposition,
the particle collection efficiency would, therefore, be 60% to
901 of that computed without accounting for condensation on par-
ticles. 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.
156
-------
Rather than 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 inertia! impaction ef-
ficiency is sufficiently high enough that all of
the 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
difference between the two for a representative case of 25% of
the volume condensing and 25% of that going to the particles
would be 5.3% (i.e., 0.25 x 0.85 x 0.25 x 100). This is a
relatively small effect compared to the other uncertainties.
Partic1e 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 deposition has been described
in EPA reports by Calvert et al. (1973, 1975, 1976). The portions
of that model relating to particle deposition were deleted to
provide a model which would describe particle growth in the ab-
sence of deposition. The basic relationships involved are as
follows:
157
-------
The rate of change of particle radius is given by a mass
balance,
where
PM =
. =
pG
PG =
at PM
2 Dr P
,. = particle to gas mass transfer
RTG dp PBM coefficient, gmol/cm2 -s -atm
water vapor partial pressure in bulk of gas,
atm
\ • ~ j
(72)
mean partial pressure to nontransferring gas, atm
particle radius, cm
gas bulk temperature, °K
molar density of water, gmol/cm3
water vapor partial pressure at vapor-liquid
interface, atm
Particle temperature can be computed from an energy balance:
where
-,
~,
= — = particle to gas heat transfer coefficient,
dp cal/cm2-s-°K
Tpi
CPP
k
LM
t
p
pG
PG
pi
(74)
= gas bulk temperature, °K
= gas Particle interface temperature, °K
= heat caPacity of particle, cal/g-°K
= thermal conductivity of gas, cal/cm2 -s-°K/cm
= latent heat of vaporization for water, cal/gmol
= time, s
= particle density, g/cm3
= particle radius
= Particle to gas mass transfer coefficient, gmol/cm2-s-atm
= water vapor partial pressure in bulk of gas, atm
= water vapor partial pressure at vapor-liquid interface, atm
158
-------
The overall energy balance for the gas-liquid interface is
given by:
k'G at LM (PG - PLi} Ap dZ =
hL at (TLi - V Ap dZ + hG at (TLi - V Ap dZ t75
where
k'r = 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
hp = heat transfer coefficient, gas to liquid,
cal/cm2-s-°K
TT = temperature of liquid bulk, °K '
LI
The equations given above can be used along with enthalpy
and material balances for the total system of gas, liquid, and
suspended particles to form a mathematical model for condensa-
tion and growth.
Prediction of Condensati.on
Calvert and Gandhi (1977) solved the equations through a
finite difference method on an electronic computer for sieve
plates under various situations to predict the fraction of
the total condensate which goes to the particles (this fraction
defined as "f "). It was found that "f " depends heavily on
"n " and liquid phase heat transfer coefficient. It decreases
significantly with "n " below about 106 particles/cm3 and does
not change much for particle number concentration greater than
107/cm3.
"f " varies between 0.1 and O.4., Calvert and Gandhi (1977)
p
used an average of 0.25 for the sieve plate scrubbers. The
fraction of the total condensate which goes to the particles
was experimentally measured to be about 0.15 for the mobile bed
159
-------
scrubber at q' * 0.25 and for water and air conditions used in
the present work. This value can be used to compute the amount
of particle growth that will result from a given condensation
ratio. If the inlet particle size distribution is known, one
can predict the overall penetration that will be achieved in
the mobile bed scrubber.
Performance Prediction Method
Case 1. Condensation and particle growth occurred within the
mobile bed scrubber.
The sequence of steps to be followed in predicting
the performance of a mobile bed F/C scrubber system is as
follows:
1. Determine the initial particle size distribution
at the scrubber inlet.
2. Compute particle penetration due to inertial im-
paction in the first stage. Use the inlet
particle size distribution and the penetration
relationship for the mobile bed given in
equation (68). In using this equation, "APW"
should be the overall pressure drop of the
scrubber, not the pressure drop for one stage.
The penetration for stage 1 is
Ptdi - Ptd1/n (76)
where Ft, = particle penetration for particle
diameter "d " in stage 1, fraction
Pt^ = overall penetration for particle
diameter "d ", fraction
n = number of stages, -
3. Calculate the condensation ratio corresponding to
the scrubber operating conditions, from this com-
pute "fv", the volume fraction of gas condensing,
160
-------
and then calculate the penetration due to diffusio-
phoresis (Ptc) according to equation (77) for a
conservative estimate or equation (78) for an
optimistic estimate:
1 - Ptc = 0.85 (fy) (l-fp) (77)
1 - Pt = 0.85 f (78)
where :
,. _ ____ mo_l£s_H_^0 conden sed _ q '
v total moles originally in" "vapor" ~ Hi + 18
where:
H! = original humidity ratio, g/g
q' = condensation ratio, g/g
The dif fusiophoretic penetration applies
equally to all particle sizes so it will not change
the size distribution but will decrease the par-
ticle concentration.
4. Determine the particle size distribution leaving
the first stage from the values of "q"' and "f "
(by equation 31 ) .
5. Compute the particle penetration for the remaining
stages of the scrubber. Use the grown particle
size distribution leaving stage 1.
Pt = Pt,n"1/n (79)
2-n d
6. The total overall fractional penetration for the
mobile bed F/C scrubber, "Ptd"> wil1 be the
product of the following:
161
-------
a) "
b) »Ptc"
c) "Pt2_n"
i.e. Ptd =
i x Ptc x Pt2_n
The above steps can be combined to obtain the following
equation:
ptd=
0.85 q' (1 - fp)
0.62 + HI
exp (-9.84 x lO'^AP l-9* d 1'6
\ w pa ,
|exp -9.84 x lO^AP
w pa2
where:
d = d
pa2 p2
0.5
1/n
(80)
(81)
772
x 10
np pp
= p
f q1 + 3.78 x 10
-10
n d 3 p
P Pi Pi
f q' + 3.78 x 10
P H
"10
n d 3 p
P Pi Pi
(82a)
(82b)
0.165 + (0.0272+f 4 d^o/p^
d = PJ_J2JL
Pi 2
0.5
(82c)
162
-------
d = grown aerodynamic particle diameter. ymA
Pa2 '
d = grown physical particle diameter, ymA
Pz
d = original aerodynamic particle diameter, ymA
£ = fraction of water vapor condensing on
particles, fraction
humidity in the saturated inlet gas, g/g
3
,
p 3.
n = particle number concentration, #/cm3
Pt, = particle penetration for diameter d , fraction
d
q' = condensation ratio, g/g
AP = overall pressure drop across mobile bed
w
scrubber, cm W.C.
p = particle density, g/cm3
n = number of mobile bed stages, --
dpi = initial Physical particle diameter, ym
p = density of grown particle
Case 2. Condensation and particle growth occurred before the
mobile bed scrubber.
The sequence of steps to be followed for this case is
similar to case 1.
1. Determine the initial particle size distribution at
the condenser inlet.
2. Calculate the condensation ratio corresponding to
the condenser operating conditions.
3. Calculate the penetration due to diffusiophoresis
according to equation (77). Collection by other
mechanisms may be neglected.
4. Calculate the grown particle size distribution at
the condenser outlet according to equations (81)
and (82).
5. Compute the particle penetration for the mobile
bed scrubber. Use the grown particle size dis-
tribution leaving the condenser.
6. Calculate the total overall fractional penetration
for the mobile bed F/C scrubber system.
163
-------
The preceding six steps are equivalent to the following
equation:
0.85 q' (1 - f )
exp (-9.84 x 10'" AP Ii96d 1>6)
w pa,, J
Pt, =
1 -
0.62 + HI
pa2
(83)
where "d " is given by equation (81), and
Pt, = particle penetration for diameter d , fraction
q' = condensation ratio, g/g
f = fraction of water vapor condensing on particles,
H! = humidity of the saturated gas at condenser inlet,
g/g
AP = overall scrubber pressure drop, cm W.C.
Overall Penetration
In order to determine the overall penetration of the system
Ptf, the penetration curve has to be integrated over the entire
range of the initial size distribution curve. This can be
accomplished either mathematically on a programmable calculator
or graphically by plotting penetration versus percent mass under -
size 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
Sample Calculation
Prediction based on the model described above was made for
case 1 and for the following conditions.
1. Three stage mobile bed with plastic net
support. Packing diameter = 3.8 cm. Bed
depth = 23 cm.
2. uQ = 2.1 cm/s
3. QL/QG =11.2 £/m3
164
-------
4. Particle size distribution at the scrubber
inlet: d = 1.6 ymA; a = 3.3
5. q' = 0.24 g/g
6. n = 5 x 107/cm3
7. TG = 70°C Ci.e. H! = 0.278 g/g)
Calculation Procedure -
Step 1: Particle size distribution at the scrubber inlet was
given.
Step 2: From equation (66), the pressure drop across the
scrubber is 15.3 cm W.C. Therefore, the particle
penetration for stage 1 is
1 6 1/3
Ptj = [exp(-0.207 d ' )] (84)
The grade penetration curve calculated from
equation (84) for the first stage is shown in
Figure 62.
Step 5: The condensation ratio, q', is 0.24 g/g. Then "fyM,
the volume fraction of gas condensing is
f = q' - °'24 - = 0.267
v H! + 18 0.278 + 1£
29
According to equation (77), the penetration due to
diffusiophoresis is
Pt = l-0.85(f )(l-fp)
c v
= 0.81
Since "Pt " is independent on particle size, the
grade penetration will be a horizontal line as
shown in Figure 62.
165
-------
o
•H
•M
u
03
f-i
O
W
2
H
cu
u
I— I
H
Pi
<
0,
1.0
0.5
0.1
0.05
0.01
I II
0.1
1. DIFFUSIOPHORETIC COLLECTION
2. STAGE 1 COLLECTION
3. STAGES 2 AND 3 COLLECTION
4. OVERALL COLLECTION
I I
0.51 5
AERODYNAMIC PARTICLE DIAMETER, ymA
10
Figure 62. Scrubber penetration for different
scrubber stages.
166
-------
Step 4: The grown aerodynamic particle size is calculated
by equations (81) and (82).
The initial and the calculated grown particle size
distribution is shown in Figure 63.
Step 5: The combined particle penetration for stages 2 and
3 are calculated from the following equation:
1-6 2/3
Pt2-3 = [exp(-0.207 d^o )]
pa 2
The calculation result is shown in Figure 62.
Step 6: The overall fractional penetration for the mobile
bed F/C scrubber is:
Pt, = 0.81 Pti Pt
2 - 3
The overall grade penetration curve for this
example is shown in Figure 62.
Comparison of Experimental Data with Predictions
The predicted F/C mobile bed performance is compared with
measurements in Figures 64 through 66. The agreement is fair.
In general, the predicted grade penetration curve crossed the
measured curve at around 1 ymA particle diameter. The model
underestimated penetration for submicron particles and over-
estimated the penetration for particles larger than 1.0 ymA
in diameter.
Figure 67 shows the predicted and measured particle pene-
tration as a function of condensation ratio for a 1 ymA dia-
meter particle. Figure 68 is a similar plot for the cut
diameter. Since the predicted and measured grade penetration
curves cross each other at 1.0 ymA, the prediction should agree
with this measurement. This is confirmed by Figure 67. The
mathematical model predicted a higher cut diameter than that
actually measured.
In the theoretical calculations, "f " is assumed to be
0.15 and n = 3 x 107/cm3 for all runs. In reality, "f " and
"n " vary from run to run. Better agreement would result if
one could use more accurate values of "f" and "n ".
167
-------
ON
CO
10
<;
z
n
o
0 1
0.5
GROWN
INITIAL
10 20 30 40 50 60 70 80 90
PERCENT BY WEIGHT UNDERSIZE, %
98
Figure 63. Initial and grown particle size distribution.
-------
c
I—I
E-
oi
UJ
OS
100
90
80
70
60
50
40
30
20
10
0
3-STAGE MOBILE BED
H = 23 cm
d, = 3.8 cm
b
Q, = 273 £/min
q' = 0..08
AP = 12.6 cm W.C.
CUPOLA DUST AEROSOL
i i i i i I
PREDICTED
0.6 0.8 1
d , ymA
Figure 64. The penetration curve for
Runs' No. FC-7 and FC-8.
169
-------
pi
tq
DH
OJ
100
90
80
70
60
50
40
30
20
10
r *
0.4
RUN NO. FC-7
= 0.11
PREDICTED
- 3-STAGE MOBILE BED
H = 23 cm W.C.
d, = 3.8 cm W.C.
b
QL = 273 i/min
AP = 12.6 cm W.C.
CUPOLA DUST AEROSOL
0.6 0.8 1
d , ymA
Figure 65. The penetration curve for Runs
No. FC-7 and FC-8.
170
-------
•P
U
rt
c
I-H
H
H
H
2
W
P-,
1.0
0.5
0.1
0. 05
0.01
I—I I i i r
PREDICTED
RUN NO. FC-23
MEASURED
a' - 0. 24
n = 3 x 107/cm3
£p = °'15
I
I I I 1
0.3 1 5 10
AERODYNAMIC PARTICLE DIAMETER, ymA
Figure 66. Predicted and measured particle
penetration for Run No. FC-23.
171
-------
o
•H
•!->
O
Oj
f-i
o
I—I
E-H
W
CL,
W
-J
U
I—i
H
C*
<
P-,
0. 5
0.1
1 1 1—1—T-TT
0.01
OHARDWARE SCREEN SUPPORT
(JPLASTIC NET SUPPORT
PREDICTED
3-STAGE MOBILE BED
n = 3 x 107/cm3
P
£ = 0.15
P
p^ = 2.2 g/cm3
AP = 13 cm W.C.
J_
J L
d =1.0 ymA
0.05 0.1
q', g vapor condensing/g dry air
0. 5
Figure 67. Predicted and measured penetration for 1.0 ymA
diameter particle versus condensation ratio.
-------
H
E-
W
o
E-
0.4
PREDICTED
3-STAGE MOBILE BED
n = 3 x 107/cm3
P
£ = 0.15
pp - 2.2 g/cm3
AP = 13 cm W.C.
CUPOLA DUST AEROSOL
I I
I I I
HARDWARE SCREEN
SUPPORT
(^PLASTIC NET
^^ SUPPORT
0.01
0.05 0.1
q', g vapor condensing/g dry gas
0.5
Figure 68. Predicted and measured variation of cut diameter
with condensation ratio.
-------
SECTION 7
EVALUATION OF MOBILE BED SCRUBBER
PARTICLE SCRUBBING
Each scrubber type has a typical cut/power relationship
which describes the dependency of scrubber performance on pressure
drop. The uniqueness of the cut/power relationship offers a
simple method to evaluate the capabilities of different scrubber
types. Figure 69 is a plot of the cut/power relationship for
several scrubber types.
Of the four scrubber types: gas atomized spray, mobile bed,
packed bed, and sieve plate, the gas atomized spray scrubber has
the highest performance capability because it can attain the
smallest cut diameter with the same pressure drop. At a pressure
drop of 20 cm W.C., the gas atomized spray scrubber can achieve
a performance cut diameter of 0.75 ymA. At the same pressure drop
the performance cut diameters are 1.6 ymA, 1.7 ymA, and 1.2 ymA
for the mobile bed, packed bed, and sieve plate with 0.32 cm hole
diameter, respectively. Industrial mobile beds usually operate at
pressure drops around 30 cm W.C. The performance cut diameter is
about 0.9 ymA which is better than packed beds and sieve plates
but less efficient than venturi scrubbers.
One can observe that at low power input the mobile bed
behaves like a sieve plate with very large holes. As power
input increases the mobile bed tends toward gas-atomized spray
performances. This is in keeping with the visual observation
that more atomization occurs as power input increases.
Other than particle collection efficiency, mobile bed scrub-
bers also have lower gas handling capacity and require more
liquid than venturi scrubbers. The superficial gas velocity in
a mobile bed ranges from 230 to 400 cm/s. The gas velocity in
174
-------
10
a:
w
H
Q
TT
3a -
T
- MOBILE BED
- PACKED BED WITH 2.5 cm
DIA. RINGS OR SADDLES
SIEVE PLATE WITH FOAM
DENSITY OF 0.4 g/cm3
AND 0.51 cm HOLE
DIAMETER
- SAME AS 3a EXCEPT
0.32 cm HOLE DIAMETER
- GAS-ATOMIZED SPRAY
10 30
PRESSURE DROP, cm W.C
100
Figure 69. Cut/power relationship for scrubbers
175
-------
the venturi throat is usually above 5000 cm/s. Therefore,
venturi scrubbers are more compact and the capital investments
are lower, although they still need a large entrainment separator,
The liquid/gas ratio in a venturi scrubber is about
2 x 10"3 cm3/cm3 (15 gal/MCF) which is about three times lower
than that in a mobile bed scrubber. This results in a higher
cost in liquid handling and moving for mobile bed scrubbers.
A mobile bed scrubber also requires a higher degree of
maintenance, especially the packing spheres. Epstein (1975)
reported that due to erosion the life of the HDPE (high denisty
polypropylene) spheres was approximately 2,000 hours. Thermo-
plastic rubber (TPR) spheres were a little better than HDPE
spheres. After 500 hours of operation, TPR spheres had lost
approximately 2.6% of their original weight and the HDPE spheres
from 8-141. After approximately 2,500 hours of operation, all
TPR spheres were dimpled on one side, about 2.4% failed at the
seam, and the weight loss average was about 6%. Epstein (1975)
estimated that the life of the TPR spheres was about one year.
The mobile bed scrubber does have one advantage over the
venturi scrubber. The mobile bed scrubber is capable of high
mass transfer and can be used to remove particulatc and
gaseous pollutants simultaneous.
MOBILE BED AS A F/C SCRUBBER
Experimental results presented in the previous sections
clearly show that the mobile scrubber can be used as a F/C
scrubber. Calvert, et al. (1975) and Calvert and Gandhi (1977)
have studied other F/C scrubber configurations - sieve plate
scrubber and spray scrubber. The mobile bed F/C scrubber is
critically evaluated and compared with the F/C sieve plate
scrubber and the F/C spray scrubber in the following sections.
Performance Capability
Mobile Bed Versus Sieve Plate Scrubber -
As discussed in the last section and revealed by Figure 69,
a non-F/C mobile bed scrubber has a higher performance capability
176
-------
than a non-F/C sieve plate scrubber when the scrubber pressure
drop is above 20 cm W.C. When operated in F/C mode, the con-
densation ratio for the mobile bed scrubber is roughly the
same as that for the sieve plate scrubber. It is expected
that the extent of particle growth will be about the same in
the two scrubber systems. Therefore, as a F/C scrubber, the
mobile bed scrubber should also have a higher performance
capability than that of a sieve plate scrubber. This is con-
firmed when comparing the results of the present study with
those reported by Calvert et al. (1975) and Calvert and Gandhi
(1977).
Calvert et al. (1975) presented results of a laboratory
pilot scale evaluation of a multiple plate sieve plate scrubber.
They plotted particle penetration against condensation ratio
for 0.6 ymA and 1.0 ymA diameter particles. The solid line
in Figure 70 shows their results for the sieve plate scrubber
with four plates. The particle number concentration was about
2 x 108/cm3 and the pressure drop across the four plates ranged
from 32 cm W.C. to 39 cm W.C. (average = 35 cm W.C.)-
In the present study, the mobile bed F/C scrubber was
operated at a pressure drop around 13 cm W.C. Therefore, the
results of the present study cannot be directly compared with
the data by Calvert et al. In the last section it was shown
that the mathematical model can reasonably predict the pene-
tration of 1 ymA diameters in a mobile bed F/C scrubber. Pre-
dictions by the mathematical model are compared with data by
Calvert et al. in Figure 70. As can be seen, a mobile bed
F/C scrubber with a pressure drop of about 30 cm W.C. will
have the same collection efficiency as a sieve plate F/C
scrubber operating at a pressure drop of 35 cm W.C.
Calvert and Gandhi (1977) reported results of a pilot
scale demonstration of F/C scrubbing for fine particle control
carried out on a secondary metal recovery furnace. The scrubber
was a 5-plate sieve plate scrubber. The scrubber performance was
reported in terms of grade penetration curves. Figure 71 shows
177
-------
c
o
u
ro
s-,
M-i
O
i— i
E-"
Pi
0.5
0.1
0.05
I I I
I
1 I I I I L
CALVERT ET AL.'S DATA FOR
4-PLATE SIEVE PLATE F/C
AP = 35 cm W.C.
n
2 x 108/cm3
4 g/cm3
PREDICTED MOBILE BED
F/C SCRUBBER PERFORMANCE
AP = 30 cm W.C.
w
d =1.0 ymA
pa
i i i i
0.01
0.05 0.1
q'» g/g
0. 5
Figure 70. Comparison between mobile F/C and
sieve plate F/C scrubber performance.
178
-------
o
•rH
+->
u
o
I—I
H
OS
W
CL,
Pi
<
P-
1.0
0.5
0.1
0.05
0.01
1 I 1 I I I I
PREDICTED MOBILE BED F/C
SCRUBBER PERFORMANCE
AP = 30 cm W.C.
w
_ AP = 35 cm W.C.
w
CALVERT AND GANDHI'S
DATA FOR 5-PLATE SIEVE
PLATE F/C
- AP = 58 cm W.C.
n = 1.9 x 108/cm3
q' = 0.31 g/g
p = 4 g/cm3
hj = 0.385 g/g
I i i I
0.1 0.51
AERODYNAMIC PARTICLE DIAMETER, ymA
Figure 71. Comparison between mobile bed
F/C and sieve plate F/C scrubber
performance.
179
-------
data for one of their runs along with the theoretical prediction
for the mobile bed scrubber. As can be seen, the mobile bed F/C
scrubber will have the same performance at a pressure drop
saving of 23 cm W.C.
Mobile Bed Versus Spray Scrubber -
Calvert et al. (1975) also reported data obtained on a hori-
zontal F/C spray scrubber. The water spraying rate was 8 £/m3
(60 gal/MCF) and the nozzle pressure was 276 kPa (40 psig). Thus,
the energy input to the spray scrubber is equivalent- to a gas
phase pressure drop of 22.4 cm W.C.
Data by Calvert et al. for a three stage cocurrent spray
are shown in Figure 72; a plot of particle penetration versus
condensation ratio for 1 umA diameter particles. In their ex-
periment, titanium dioxide was used as the test, aerosol. The
particle number concentration was quite low, approximately
2 x 106 - 6 x 106/cm3.
The predicted mobile bed F/C scrubber performance with a
scrubber pressure drop of 22.4 cm W.C. and a particle number
concentration of 1 x 107/cm3 and 5 x 106/cm3 is plotted in
Figure 72. The predicted mobile bed performance is slightly
better than the measured performance of the spray scrubber.
Cost Comparison -
Capital cost - The cost advantage of F/C scrubber systems
over the conventional high energy scrubber systems has been
demonstrated by Calvert et al. (1975) and Calvert and Gandhi
(1977). This comparison will not be repeated here., In the
following sections, the relative costs between a sieve plate
F/C, a mobile bed F/C, and a spray F/C system are presented.
Although the details of the F/C scrubbing system design
will be different for each source, the overall process design,
illustrated in Figure 73, will be the same. The major com-
ponents of the system includes a quencher, a F/C scrubber,
and a cooling tower.
180
-------
1.0
0.5
n
o
• H
u
TO
£
z
o
hH
H °' *
W
2
PL,
UJ
j 0.05
u
H
<
PH
0.01
: 1 i — i MUM 1 1 — rn
- PREDICTED MOBILE BED F/C
SCRUBBER PERFORMANCE
APW= 22.4 cm W.C.
p = 4 g/cm3
P
— _
x ,^-n = 1 x 107/cm
"^^o*1^^ P
^^!s CALVERT ET AL . DATA"
X. FOR SPRAY F/C
x ^^n =lxl06-6x!06/cm3
/ \ ^. /
7 \ \ X /
n =5xl06/cm3 \ \X/ I
\ \ \
* V
\ \
\
V
\
\
d =1.0 ymA \
pa \^
\
1 1 1 1 1 1 1 1 II
0.01
0.05 0.1
q', g/g
0. 5
Figure 72.
Comparison between mobile bed
F/C and spray F/C scrubber
performance.
181
-------
TO FAN
AND STACK
AIR
HOT GAS
FROM SOURCE
i
QUEN-
CHER
1
\
t
\
I
F/C
SCRUBBER
t
r
CAUSTIC FOR pH
CONTROL
!
LIQUID
TREATMENT
i
COOLING
TOWER
I
AIR
DT
MAKE-UP
"•—WATER
AND SLUDGE
TO DRAIN
Figure 73. Typical process diagram of a F/C
scrubber system.
182
-------
The costs of the quencher, piping and ductwork, liquid
treatment, instrumentation, and process control are roughly
the same for the three F/C scrubber systems. The major cost
differences between the three F/C scrubber systems are due to
the costs of the scrubber itself and the cooling tower. This
is because the mobile bed F/C scrubber system has a higher gas
handling capacity than the sieve plate and spray scrubber. How-
ever, the mobile bed requires more liquid which results in a
larger cooling tower.
To analyze the relative costs, the F/C scrubber system is
assumed to be for cupola gas cleaning. The volumetric gas flow
rate at the scrubber inlet is 1,300 Am3/min (45,900 ACFM). Op-
erating conditions for the three scrubber systems are as follows:
Sieve plate: ufi = 150 cm/s (5 ft/secj
QL/QG = 4 £/min (30 gal/MCF)
AP = 35 cm W.C.
Spray Scrubber: un = 110 cm/s (3.6 ft/sec)
\j
QL/QG = 8 i/m3 (60 gal/MCF)
AP = negligible
Spray nozzle pressure = 276 kPa (40 psig)
Mobile bed: un = 340 cm/s (11 ft/secj
b
QL/QG = 8 A/min (60 gal/MCF)
AP = 25 cm W.C.
Based on these conditions, the estimated fabrication costs
of the three scrubbers are $25,000 (0.54/CFM), $20,000 (0.44/CFM),
and $21,000 (0.46/CFM) for the sieve plate scrubber, mobile bed
scrubber and the spray scrubber, respectively. The above costs
only include material cost and fabrication labor cost.
Costs for design, administration, contingency, etc. are not
included since those costs are about the same for the three
scrubbers. Since the installed cost is about four times the
fabrication cost, the cost of the sieve plate scrubber is approx-
imately $20,000 more than the mobile bed and spray scrubbers.
The price quotation for the cooling towers from the manufac-
turer is $40,000 for the cooling towers in the mobile bed scrubber
183
-------
system and in the spray scrubber system. It is $20,000 for the
cooling tower in the sieve plate system. The quoted price in-
cludes the costs of the cooling tower and fan, but it did
not include the costs of the pump and the electrical connec-
tions .
Based on this analysis, it appears that the capital cost
for the three F/C scrubber systems will be approximately the
same .
Operating cost - If the scrubber and the cooling tower
can be installed close together, the power requirement to
operate the scrubber and cooling tower is 375 hp, 310 hp, and
295 hp for the mobile bed, sieve plate, and spray scrubber, re-
spectively. A fan efficiency of 50% and a pump efficiency of
65% were assumed in the calculation. The mobile bed F/C system
has the highest power requirement. The annual operating cost
(not including annualized capital charges and depreciation) for
the mobile bed F/C system is approximately $19,000 higher than
that of the sieve plate F/C system and approximately $23,000
higher than the spray F/C scrubber system.
Maintenance - There will be no unusual maintenance problems
with the mobile bed and sieve plate scrubber systems even though
the mobile bed packing might have to be replaced every year. This
is not the case with the spray scrubber. In order to obtain high
collection efficiency for particles, the liquid drops should be
small in diameter. In the experimental study reported by Calvert
et al. (1975), drops as small as 400 ym in diameter were used.
Spray nozzles which can produce this drop size requires high
pressure. The liquid flow rate per nozzle is small, less than
1 GPM. Therefore, a large quantity of spray nozzles are required.
The manpower requirement to maintain the proper operation of the
nozzles will be great.
POTENTIAL FOR POWER PLANT APPLICATION
Mobile bed scrubbers have been used in coal-fired power
plants for the control of S0x emissions as well as particulate
184
-------
emissions. They are sometimes installed after the electrostatic
precipitators as secondary collectors and this practice is
expected to become more prevalent. At the precipitator outlet
flue gas temperature is about 138°C (280°F) and contains about
10% by volume of moisture. Typical fly ash size distribution
at the ESP outlet is d = 3.0 ymA and a = 3.0 (Figure 74).
Particle concentration is about 0.5 g/DNm3 (0.2 gr/SCF).
A three-stage mobile bed scrubber without F/C effect and
operated at a pressure drop of 30 cm W.C. would have a grade
penetration curve as that shown in Figure 75. The predicted
overall particle penetration will be 17% (overall collection
efficiency = 831) and the predicted outlet particle loading will
be 0.085 g/DNm3 (0.034 gr/SCF).
Assume the flue gas can be saturated by evaporation of water;
it will saturate around 54°C (130°F) and the moisture content will
be 0.11 g/g. By assuming that the particle number concentration
is 1 x 108/cm3 and the flue gas is cooled down to 43°C (110°F),
a three-stage F/C mobile bed scrubber with a pressure drop of
30 cm W.C. will have a grade penetration curve like that shown by
the dashed line in Figure 75. The predicted overall penetration
will be 12% (88% overall collection efficiency) and the outlet
particle loading will be 0.06 g/DNm3 (0.024 gr/SCF).
For this application, the improvement in collection effi-
ciency of the F/C mobile bed scrubber over the non-F/C mobile
bed is small. The main reason is that the condensation ratio
attainable in the power plant is too small (-0.056 g/g). If a
large quantity of waste steam is available, the condensation
ratio may be increased by injecting the steam into the gas.
Steam injection causes an extremely high saturation ratio in
the vicinity of the injection nozzle and this enhances the
nucleation of condensation and particle growth. However, even
though waste steam may be available, the mobile bed scrubber
might not be a good choice for a F/C scrubber since the mobile
bed is not an efficient particle scrubber. A better choice could
ne the combination of spray quencher and venturi scrubbers. The
185
-------
10
<
HH
Q
W
i—i
PC:
u
i—i
2-
^
o
0.3
I I
I I
np = 1 x 108/g
Pp = 2.2 g/cm3
q' = 0.056 g/g
I
GROWN SIZE
ORIGINAL SIZE
DISTRIBUTION
till
0.2 12 5 10 20 30 40 50 60 70 80
PERCENT BY WEIGHT UNDERSIZE, I
Figure 74. Typical fly ash distribution and
grown size distribution.
186
-------
o
•H
•P
O
rt
2
O
H
W
2
W
PH
w
_)
u
I—I
H
1.0
0.5 -
0.1 _
0.05 -
0.01
MOBILE BED
F/C MOBILE BED
np = 1 x 10a/g
q' = 0.056 g/g
P = 2.2 g/cm3
SCRUBBER PRESSURE DROP = 30 cm W.C.
I I I I I I I l l i I
0.1 0.51 5
AERODYNAMIC PARTICLE DIAMETER, ymA
Figure 75. Predicted mobile bed and F/C
mobile bed scrubber performance.
187
-------
spray section is used for SO collection and for condensation
and particle growth.
MOBILE BED SCRUBBER DESIGN RECOMMENDATIONS
The particle collection efficiency of a mobile bed scrubber
depends only on the pressure drop across the bed. McMichael
et al. (1976) and Wen and Chang (1978) have shown that the collec-
tion efficiency of a mobile bed scrubber for sulfur dioxide is
also dependent only on the pressure drop. Therefore, if the
pressure drop across the bed can be raised, the particle and
S02 collection efficiencies will increase.
As mentioned earlier, most of the mobile bed scrubber pres-
sure drop is due to the weight of the packings, liquid holdup in
the bed, and the liquid head retained on the supporting grids.
An increase in any one of these will cause the scrubber pressure
drop to increase.
The weight of packing in a mobile bed may be increased by
increasing the static bed height by using heavier packings, and
by using smaller packings. The packing spheres commonly used in
industrial mobile bed scrubbers are hollow 3.8 cm diameter poly-
propylene balls. The average weight for each ball is about 4.5 g
- 5 g (packing density = 0.137 - 0.175 g/cm3). Recently, the
scrubber manufacturers have recommended using heavier balls.
Balls as heavy as 7 g (packing density = 0.244 kg/cm3) were
recommended.
Increasing the packing density has an added advantage: heat
and mass transfer rates will be higher. According to O'Neill et
al. (1972), the mobile bed can be operated either in the mode of
fluidization without flooding or in the mode of fluidization due
to incipient flooding. The operation mode of fluidization due
to incipient flooding is the preferable mode. The flooding
operation mode results in a higher liquid holdup in the bed and
a higher gas-liquid interfacial activity which gives higher heat
and mass transfer rate.
188
-------
O'Neill et al. (1972) also say that the mode of operation de-
pends largely on packing density and to a lesser extent on the
packing size, liquid flow rate and physical properties of the
liquid. Based on Chen and Douglas's empirical equation for liquid
holdup, O'Neill et al. calculated that the transition of the non-
flooding mode to the flooding mode occurred at a packing density of
200 kg/m3 for the 3.8 cm diameter packing. Thus, industrial
mobile beds are operated in the nonflooding mode. The use of
heavier 7 g spheres will shift the operation mode from nonflooding
to flooding,
Smaller balls of the same wall thickness have a higher
packing density. The use of smaller packing also causes liquid
holdup in the bed to increase and causes the bed to be operated
in the flooding mode sooner. The increase in weight and liquid
holdup increases the pressure drop across the bed and improves
the particle collection efficiency.
Liquid head retained on the supporting grids may be in-
creased by using grids with smaller openings and a smaller
percentage of open area. For a supporting grid with small
openings and open area, it acts as a sieve tray operating in
the weeping region. Liquid cannot freely flow downward through
the openings and are partially retained on the tray. As in a
sieve column, the liquid retained on the tray contributes to
particle collection and pressure drop.
Another advantage of using supporting grids with small
openings is that the bed motion is more uniform. The bed
motion of the mobile bed with plastic net supports was ob-
served to be more uniform than the bed with hardware screen
supports., This observation conforms with others' findings.
Numerous researchers have indicated that in the fluidization
of solid by gas, the pressure drop across the gas distribution
plate should be high enough to give uniform fluidization. The
mobile bed is an irrigated fluidization of solid by gas. Thus
to obtain uniform fluidization, the pressure drop across the
189
-------
supporting grid should be high. One method to obtain high
pressure drop across the grids is to use grids with less
percentage of open area as well as smaller openings.
190
-------
SECTION 8
FUTURE RESEARCH RECOMMENDATIONS
The objectives of studying the particle collection in a
mobile bed scrubber and determining the feasibility of using
it as a F/C scrubber have been achieved in this study through
experimental and theoretical evaluation. It has been clearly
shown that mobile beds are capable of being used as F/C scrubbers
In order for the mobile bed scrubbers to be successfully
used as F/C scrubbers, some fundamental research and development
work are required. We recommend future research work in the
following areas:
1. Theoretical and experimental study of conden-
sation due to SO sorption on the fly ash.
2C
2. Experimental and theoretical determination of
the specific details of heat and mass transfer
in mobile bed scrubbers.
3. Development of cooling devices suited for the
cooling of scrubber liquid containing suspended
and dissolved solids.
4. Experimental and theoretical evaluation of the
effects of steam injection.
5. Experimental and theoretical study of particle
growth due to condensation.
F/C Effects from SO Sorption of Fly Ash
When fuel containing sulfur is burned, the hydrocarbons
present are converted to sulfur dioxide. Part of the sulfur
dioxide will oxidize further to sulfur trioxide at high temper-
ature. Sulfur trioxide exists in dissociated form at high
temperature. As the flue gas mixture is cooled, sulfur tri-
191
-------
oxide associates with water vapor to form sulfuric acid vapor.
Part of the acid vapor will be adsorbed by the fly ash.
As the gas cools down to below the dew point temperature,
such as inside a scrubber, the remaining acid vapor will con-
dense as acid mist. Due to the strong affinity of sulfuric
acid towards water, the acid will absorb water. For those fly
ash particles with acid adsorbed on their surfaces, the mass
will increase. This phenomenon has the same effect as particle
growth due to condensation of water vapor.
Mobile bed scrubbers are usually installed after the electro-
static precipitator as an additional particle scrubber. To
decrease the fly ash resistivity, some ESP users use sulfur
trioxide as the conditioning agent by injecting the S03 gas into
the flue gas. Therefore, it is possible to use S03 as the con-
ditioning agent for the electrostatic precipitator and as the
F/C agent for the scrubber systems after the ESP. We recommend
that this possibility be fully evaluated to determine its
applicability. The following approach is recommended.
1. Survey literature on the adsorption of S03 by fly ash,
the nucleation of sulfuric acid, and the absorption of
water by the sulfuric acid.
2. Conduct a bench scale study to obtain information on the
conditions under which particle growth will occur and
the rate of particle growth.
3. Determine the technical and economic feasibilities of
using S03 as the F/C agent.
4. Select promising scrubber systems and conduct a detailed
experimental study as follows:
a) Laboratory pilot scale study with scrubber capacity
of 28 mVmin (1,000 CFM) .
b) Pilot scale demonstration in a power plant with
scrubber capacity around 280 m3/min (10,000 CFM).
c) Full scale demonstration in a power plant.
192
-------
Heat and Mass Transfer in Mobile Bed Scrubbers
Information on the rates of heat and mass transfer and
the gas-liquid contacting area are required for the proper
design of mobile bed F/C scrubbers. The mass transfer rate
and contacting area also determine the absorption efficiency
of SOa in lime and limestone slurry.
A few empirical correlations are available in the liter-
ature for the predictions of contacting area and heat and mass
transfers; e.g., Wofniak (1975) on gas-liquid interfacial area,
Gel'perin, et al. (1973) on heat and mass transfers, McMichael,
et al. (1976) and Wen and Chang (1977) on absorption of S02
in lime and limestone slurry. The application of these cor-
relations are limited to scrubber geometries and conditions
similar to those used by these investigators. The adequacy of
these correlations for other situations has not been established
Therefore, it would be beneficial to study the specific details
of heat and mass transfer in mobile bed scrubbers both theoretic-
ally and experimentally. The following approach is recommended:
1. Review and assess the published literature and
available unpublished information pertaining to
interfacial area and heat and mass transfer in
mobile bed scrubbers.
2. Determine all factors which affect the interfacial
area and heat and mass transfer rates and develop
theory.
3. Conduct experiments to verify or improve the
theory.
Development of Liquid Cooling System
Due to the large requirement of cold scrubber liquor and
the complications introduced by dissolved and suspended solids,
the liquor cooling procedure has a significant effect on the
economics of an F/C scrubber system. It represents the major
cost difference between F/C and conventional scrubbing for
many applications.
193
-------
Concern over the possibility that solids deposition on
cooling tower surfaces would lead to heavy buildup of adherent
scale led to the choice of a spray-type cooler in the present
study. The spray coolers have some drawbacks, such as low
efficiency, high pressure nozzles required to produce small
drops, and the plugging of spray nozzles. Therefore, the pos-
sibility of using packed or filled cooling towers should be
investigated. The use of a standard commercial cooling tower
would give the best combinations 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.
We recommend studies of solids deposition on surfaces of
various materials which could be used for cooling tower con-
struction. 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.
Steam Injection
Mobile bed scrubbers are currently used in power plants for
removing particulates and S02 from flue gas by utilizing lime
and limestone slurries. The gas temperature and vapor content
of the power plant are generally low (280°F, 10% by volume
moisture). Therefore, unless large quantities of spent steam
are available, condensation effects and particle growth, would be
minimal.
The use of steam injection into saturated gas is an attrac-
tive 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 idenfity
the best way of generating steam inexpensively.
Data from previous studies indicated that fine particle col-
lection efficiency was greater when a given amount of steam was
introduced into the gas than when an equivalent quantity was
194
-------
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 conden-
sation 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 in-
jection is (are) not known but can be hypothesized. If steam
is mixed with saturated gas, only a small fraction will 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 sig-
nificant for insoluble particles than for soluble ones.
If it is possible to obtain a substantial benefit from in-
jection 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 propor-
tions 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 engineer-
ing analysis and design studies. Conventional and nonconventional
waste heat boilers should be evaluated. Cooling costs can be
based on existing technology.
195
-------
Particle Growth
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 nonwettable 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 pheno-
mena 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 pre-
sently 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. 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 fraction
of total condensation which goes to the particles (i.e., "f ")
as predicted by the mathematical model.
196
-------
3. The particle number concentration also influences "£ "
P
as computed from the model. The experimental data taken pre-
viously have not enabled very precise computation of number
concentration, so the influence of this parameter has not been
distinguishable from those of other parameters.
4. Soluble particles can cause condensation at saturation
ratios less than 1.0 because the vapor pressure of water is
lowered by the solute.
197
-------
REFERENCES
Balabekov, V. S., P. G. Romankov, E. Ya. Tarat, and M. F.
Mikhalev. J. Appl. Chem. USSR, 4_2:1454, 1969.
Balabekov, V. S., E. Ya. Tarat, P. G. Romankov, and M. F.
Mikhalev. J. Appl. Chem. USSR, £4:1061, 1971.
Barile, R. G. and D. W. Meyer. Turbulent Bed Cooling Tower.
Chem. Eng. Progr. Symposium Series 67, No. 119, 1971.
pp. 134-143.
Blyakher, I. G., L. Ya. Zhivaikin, and N. A. Yurovskaya.
Investigation of Hydrodynamics and Mass Transfer in
Equipment with Movable Packing. Int'l Chem. Engr.,
7_:485-490, 1967.
Borgwardt, R. Limestone Scrubbing of S02 at EPA Pilot Plant.
Report No. 1. August 1972.
Borgwardt, R. Limestone Scrubbing of S02 at EPA Pilot Plant.
Report No. 14. January 1974a
Borgwardt, R. Limestone Scrubbing of S02 at EPA Pilot Plant.
Report No. 16. June 1974b.
Borgwardt, R. Limestone Scrubbing of S02 at EPA Pilot Plant.
Report No. 21. June 1975.
Calvert, S. Source Control by Liquid Scrubber. In: Air Pol-
lution, Chapter 46, Arthur Stern, ed. Academic Press,
New York, 1968.
Calvert, S., J. Goldshmid, D. Leith, and D. Mehta. Wet Scrubber
System Study, Volume 1, Scrubber Handbook. EPA-R2-72-118a,
NTIS PB 213-016, August 1972.
Calvert, S., J. Goldshmid, D. Leith, and N. Jhaveri. Feasi-
bility of Flux Force/Condensation Scrubbing for Fine
Particulate Collection. EPA 650/2-73-036, NTIS PB 227-
307, 1973.
Calvert, S. Engineering Design of Fine Particle Scrubbers.
J. of A.P.C.A., 24:929-933, 1974.
198
-------
emissions. They are sometimes installed after the electrostatic
precipitators as secondary collectors and this practice is
expected to become more prevalent. At the precipitator outlet,
flue gas temperature is about 138°C (280°F) and contains about
10% by volume of moisture. Typical fly ash size distribution
at the ESP outlet is d = 3.0 ymA and a = 3.0 (Figure 74).
Particle concentration is about 0.5 g/DNm3 (0.2 gr/SCF).
A three-stage mobile bed scrubber without F/C effect and
operated at a pressure drop of 30 cm W.C. would have a grade
penetration curve as that shown in Figure 75. The predicted
overall particle penetration will be 17% (overall collection
efficiency = 83%) and the predicted outlet particle loading will
be 0.085 g/DNm3 (0.034 gr/SCF).
Assume the flue gas can be saturated by evaporation of water;
it will saturate around 54°C (130°F) and the moisture content will
be 0.11 g/g. By assuming that the particle number concentration
is 1 x 10B/cm3 and the flue gas is cooled down to 43°C (110°F)5
a three-stage F/C mobile bed scrubber with a pressure drop of
30 cm W.C. will have a grade penetration curve like that shown by
the dashed line in Figure 75. The predicted overall penetration
will be 12% (88% overall collection efficiency) and the outlet
particle loading will be 0.06 g/DNm3 (0.024 gr/SCF),
For this application, the improvement in collection effi-
ciency of the F/C mobile bed scrubber over the non-F/C mobile
bed is small. The main reason is that the condensation ratio
attainable in the power plant is too small (-0,056 g/g). If a
large quantity of waste steam is available, the condensation
ratio may be increased by injecting the steam into the gas.
Steam injection causes an extremely high saturation ratio in
the vicinity of the injection nozzle and this enhances the
nucleation of condensation and particle growth. However,, even
though waste steam may be available, the mobile bed scrubber
might not be a good choice for a F/C scrubber since the mobile
bed is not an efficient particle scrubber. A better choice could
be the combination of spray quencher and venturi scrubbers. The
185
-------
10
E-
S
i—i
O
_J
I-H
E-
O
o
OS
w
0.3
np = 1 x 108/g
p = 2. 2 g/cm3
q1 - 0.056 g/g
GROWN SIZE
ORIGINAL SIZE
DISTRIBUTION
I l l I
0.2 125 10 20 30 40 50 60 70 80
PERCENT BY WEIGHT UNDERSIZE, %
Figure 74. Typical fly ash distribution and
grown size distribution.
186
-------
n
o
o
03
VH
M-i
2;
o
w
CH
W
1. 0
0.5
0.1
0.05
0. 01
F/C MOBILE BED
np = 1 x 10B/g
q' = 0.056 g/g
pp = 2.2 g/cm3
SCRUBBER PRESSURE DROP = 30 cm W.C.
I I I I | | | i i i
0.1 0.5 1
AERODYNAMIC PARTICLE DIAMETER,
Figure 75. Predicted mobile bed and F/C
mobile bed scrubber performance.
187
-------
spray section is used for SO collection and for condensation
JC
and particle growth.
MOBILE BED SCRUBBER DESIGN RECOMMENDATIONS
The particle collection efficiency of a mobile bed scrubber
depends only on the pressure drop across the bed. McMichael
et al. (1976) and Wen and Chang (1978) have shown that the collec-
tion efficiency of a mobile bed scrubber for sulfur dioxide is
also dependent only on the pressure drop. Therefore, if the
pressure drop across the bed can be raised, the particle and
S02 collection efficiencies will increase.
As mentioned earlier, most of the mobile bed scrubber pres-
sure drop is due to the weight of the packings, liquid holdup in
the bed, and the liquid head retained on the supporting grids.
An increase in any one of these will cause the scrubber pressure
drop to increase.
The weight of packing in a mobile bed may be increased by
increasing the static bed height by using heavier packings, and
by using smaller packings. The packing spheres commonly used in
industrial mobile bed scrubbers are hollow 3.8 cm diameter poly-
propylene balls. The average weight for each ball is about 4.5 g
- 5 g (packing density = 0.137 - 0.175 g/cm3). Recently, the
scrubber manufacturers have recommended using heavier balls.
Balls as heavy as 7 g (packing density = 0.244 kg/cm3) were
recommended.
Increasing the packing density has an added advantage: heat
and mass transfer rates will be higher. According to O'Neill et
al. (1972), the mobile bed can be operated either in the mode of
fluidization without flooding or in the mode of fluidization due
to incipient flooding. The operation mode of fluidization due
to incipient flooding is the preferable mode. The flooding
operation mode results in a higher liquid holdup in the bed and
a higher gas-liquid interfacial activity which gives higher heat
and mass transfer rate.
188
-------
O'Neill et al. (1972) also say that the mode of operation de-
pends largely on packing density and to a lesser extent on the
packing size, liquid flow rate and physical properties of the
liquid. Based on Chen and Douglas's empirical equation for liquid
holdup, O'Neill et al. calculated that the transition of the non-
flooding mode to the flooding mode occurred at a packing density of
200 kg/m3 for the 3,8 cm diameter packing. Thus, industrial
mobile beds are operated in the nonflooding mode. The use of
heavier 7 g spheres will shift the operation mode from nonflooding
to flooding.
Smaller balls of the same wall thickness have a higher
packing density. The use of smaller packing also causes liquid
holdup in the bed to increase and causes the bed to be operated
in the flooding mode sooner. The increase in weight and liquid
holdup increases the pressure drop across the bed and improves
the particle collection efficiency.
Liquid head retained on the supporting grids may be in-
creased by using grids with smaller openings and a smaller
percentage of open area. For a supporting grid with small
openings and open area, it acts as a sieve tray operating in
the weeping region. Liquid cannot freely flow downward through
the openings and are partially retained on the tray. As in a
sieve column, the liquid retained on the tray contributes to
particle collection and pressure drop.
Another advantage of using supporting grids with small
openings is that the bed motion is more uniform. The bed
motion of the mobile bed with plastic net supports was ob-
served to be more uniform than the bed with hardware screen
supports. This observation conforms with others' findings.
Numerous researchers have indicated that in the fluidization
of solid by gas, the pressure drop across the gas distribution
plate should be high enough to give uniform fluidization. The
mobile bed is an irrigated fluidization of solid by gas. Thus
to obtain uniform fluidization, the pressure drop across the
189
-------
supporting grid should be high. One method to obtain high
pressure drop across the grids is to use grids with less
percentage of open area as well as smaller openings.
190
-------
SECTION 8
FUTURE RESEARCH RECOMMENDATIONS
The objectives of studying the particle collection in a
mobile bed scrubber and determining the feasibility of using
it as a F/C scrubber have been achieved in this study through
experimental and theoretical evaluation. It has been clearly
shown that mobile beds are capable of being used as F/C scrubbers
In order for the mobile bed scrubbers to be successfully
used as F/C scrubbers, some fundamental research and development
work are required. We recommend future research work in the
following areas:
1. Theoretical and experimental study of conden-
sation due to SO sorption on the fly ash.
.X.
2. Experimental and theoretical determination of
the specific details of heat and mass transfer
in mobile bed scrubbers.
3. Development of cooling devices suited for the
cooling of scrubber liquid containing suspended
and dissolved solids.
4. Experimental and theoretical evaluation of the
effects of steam injection.
5. Experimental and theoretical study of particle
growth due to condensation.
F/C Effects from S0x Sorption of Fly Ash
When fuel containing sulfur is burned, the hydrocarbons
present are converted to sulfur dioxide. Part of the sulfur
dioxide will oxidize further to sulfur trioxide at high temper-
ature. Sulfur trioxide exists in dissociated form at high
temperature. As the flue gas mixture is cooled, sulfur tri-
191
-------
oxide associates with water vapor to form sulfuric acid vapor.
Part of the acid vapor will be adsorbed by the fly ash.
As the gas cools down to below the dew point temperature,
such as inside a scrubber, the remaining acid vapor will con-
dense as acid mist. Due to the strong affinity of sulfuric
acid towards water, the acid will absorb water. For those fly
ash particles with acid adsorbed on their surfaces, the mass
will increase. This phenomenon has the same effect as particle
growth due to condensation of water vapor.
Mobile bed scrubbers are usually installed after the electro-
static precipitator as an additional particle scrubber. To
decrease the fly ash resistivity, some ESP users use sulfur
trioxide as the conditioning agent by injecting the 863 gas into
the flue gas. Therefore, it is possible to use S03 as the con-
ditioning agent for the electrostatic precipitator and as the
F/C agent for the scrubber systems after the ESP. We recommend
that this possibility be fully evaluated to determine its
applicability. The following approach is recommended.
1. Survey literature on the adsorption of S03 by fly ash,
the nucleation of sulfuric acid, and the absorption of
water by the sulfuric acid.
2. Conduct a bench scale study to obtain information on the
conditions under which particle growth will occur and
the rate of particle growth.
3. Determine the technical and economic feasibilities of
using S03 as the F/C agent.
4. Select promising scrubber systems and conduct a detailed
experimental study as follows:
a) Laboratory pilot scale study with scrubber capacity
of 28 m3/min (1,000 CFM).
b) Pilot scale demonstration in a power plant with
scrubber capacity around 280 m3/min (10,000 CFM).
c) Full scale demonstration in a power plant.
192
-------
Heat and Mass Transfer in Mobile Bed Scrubbers
Information on the rates of heat and mass transfer and
the gas-liquid contacting area are required for the proper
design of mobile bed F/C scrubbers. The mass transfer rate
and contacting area also determine the absorption efficiency
of S02 in lime and limestone slurry.
A few empirical correlations are available in the liter-
ature for the predictions of contacting area and heat and mass
transfers; e.g., Wozniak (1975) on gas-liquid interfacial area,
Gel'perin, et al. (1973) on heat and mass transfers, McMichael,
et al. (1976) and Wen and Chang (1977) on absorption of S02
in lime and limestone slurry. The application of these cor-
relations are limited to scrubber geometries and conditions
similar to those used by these investigators. The adequacy of
these correlations for other situations has not been established
Therefore, it would be beneficial to study the specific details
of heat and mass transfer in mobile bed scrubbers both theoretic-
ally and experimentally. The following approach is recommended:
1. Review and assess the published literature and
available unpublished information pertaining to
interfacial area and heat and mass transfer in
mobile bed scrubbers.
2. Determine all factors which affect the interfacial
area and heat and mass transfer rates and develop
theory.
3. Conduct experiments to verify or improve the
theory.
Development of Liquid Cooling System
Due to the large requirement of cold scrubber liquor and
the complications introduced by dissolved and suspended solids,
the liquor cooling procedure has a significant effect on the
economics of an F/C scrubber system. It represents the major
cost difference between F/C and conventional scrubbing for
many applications.
193
-------
Concern over the possibility that solids deposition on
cooling tower surfaces would lead to heavy buildup of adherent
scale led to the choice of a spray-type cooler in the present
study. The spray coolers have some drawbacks, such as low
efficiency, high pressure nozzles required to produce small
drops, and the plugging of spray nozzles. Therefore, the pos-
sibility of using packed or filled cooling towers should be
investigated. The use of a standard commercial cooling tower
would give the best combinations 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.
We recommend studies of solids deposition on surfaces of
various materials which could be used for cooling tower con-
struction. 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.
Steam Injection
Mobile bed scrubbers are currently used in power plants for
removing particulates and S02 from flue gas by utilizing lime
and limestone slurries. The gas temperature and vapor content
of the power plant are generally low (280°F, 10% by volume
moisture). Therefore, unless large quantities of spent steam
are available, condensation effects and particle growth would be
minimal.
The use of steam injection into saturated gas is an attrac-
tive 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 idenfity
the best way of generating steam inexpensively.
Data from previous studies indicated that fine particle col-
lection efficiency was greater when a given amount of steam was
introduced into the gas than when an equivalent quantity was
194
-------
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 conden-
sation 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 in-
jection is (are) not known but can be hypothesized. If steam
is mixed with saturated gas, only a small fraction will 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 sig-
nificant for insoluble particles than for soluble ones.
If it is possible to obtain a substantial benefit from in-
jection 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 propor-
tions 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 engineer-
ing analysis and design studies. Conventional and nonconventional
waste heat boilers should be evaluated. Cooling costs can be
based on existing technology.
195
-------
Particle Growth
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 nonwettable 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 pheno-
mena 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 pre-
sently 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. 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 fraction
of total condensation which goes to the particles (i.e., "f ")
as predicted by the mathematical model.
196
-------
3. The particle number concentration also influences "f ",
as computed from the model. The experimental data taken pre-
viously have not enabled very precise computation of number
concentration, so the influence of this parameter has not been
distinguishable from those of other parameters.
4. Soluble particles can cause condensation at saturation
ratios less than 1.0 because the vapor pressure of water is
lowered by the solute.
197
-------
REFERENCES
Balabekov, V. S., P. G. Romankov, E. Ya. Tarat, and M. F.
Mikhalev. J. Appl. Chem. USSR, 4_2:1454, 1969.
Balabekov, V. S., E. Ya. Tarat, P. G. Romankov, and M. F.
Mikhalev. J. Appl. Chem. USSR, 44:1061, 1971.
Barile, R. G. and D. W. Meyer. Turbulent Bed Cooling Tower.
Chem. Eng. Progr. Symposium Series 67, No. 119, 1971.
pp. 134-143.
Blyakher, I. G., L. Ya. Zhivaikin, and N. A. Yurovskaya.
Investigation of Hydrodynamics and Mass Transfer in
Equipment with Movable Packing. Int'l Chem. Engr.,
7^:485-490, 1967.
Borgwardt, R. Limestone Scrubbing of S02 at EPA Pilot Plant.
Report No. 1. August 1972.
Borgwardt, R. Limestone Scrubbing of S02 at EPA Pilot Plant.
Report No. 14. January 1974a
Borgwardt, R. Limestone Scrubbing of S02 at EPA Pilot Plant.
Report No. 16. June 1974b.
Borgwardt, R. Limestone Scrubbing of S02 at EPA Pilot Plant.
Report No. 21. June 1975.
Calvert, S. Source Control by Liquid Scrubber. In: Air Pol-
lution, Chapter 46, Arthur Stern, ed. Academic Press,
New York, 1968.
Calvert, S., J. Goldshmid, D. Leith, and D. Mehta. Wet Scrubber
System Study, Volume 1, Scrubber Handbook. EPA-R2-72-118a,
NTIS PB 213-016, August 1972.
Calvert, S., J. Goldshmid, D. Leith, and N. Jhaveri. Feasi-
bility of Flux Force/Condensation Scrubbing for Fine
Particulate Collection. EPA 650/2-73-036, NTIS PB 227-
307, 1973.
Calvert, S. Engineering Design of Fine Particle Scrubbers.
J. of A.P.C.A., 24.:929-933, 1974.
198
-------
Calvert, S., N. Jhaveri, and T. Huiskong. Study of Flux Force/
Condensation Scrubbing of Fine Particles. EPA 600/2-75-018,
NTIS PB 249-297, 1975.
Calvert, S. and S. Gandhi. Fine Particle Collection by a Flux
Force/Condensation Scrubber: Pilot Demonstration. EPA
600/2-77-238, NTIS PB 227-075, 1977.
Calvert, S., S. Yung, and L. E. Sparks. Liquid Entrainment from
a Mobile Bed Scrubber. J. of A.P.C.A., 2^_:768-770, 1977.
Chen, B. H. and W. J. M. Douglas. Axial Mixing of Liquid in
a Turbulent-Bed Contactor. Can. J. Chem. Engr. 47:113-118,
April 1969.
Chen, B. H. and W. J. M. Douglas. Liquid Hold-up and Minimum
Fluidization Velocity in a Turbulent Contactor. Can. J.
Chem. Engr., 4_6:245-249, August 1968.
Douglas, H. R., I. W. A. Snider, and G. Tomlinson. The Tur-
bulent Contact Absorber. Chem. Engr. Prog., 59:85-89,
December 1963.
Douglas, W. J. M. Heat and Mass Transfer in a Turbulent Bed
Contactor. Chem. Eng. Progr., 6_0_: 66-71, July 1964.
Ensor, D. S., B. S. Jackson, S. Calvert, C. Lake, D. V. Wallon,
R. E. Nilan, K. S. Campbell, T. A. Cahill, and R. G.
Flocchini. Evaluation of a Particulate Scrubber on a
Coal-Fired Utility Boiler. EPA 600/2-75-074, November
1975.
Epstein, M. EPA Alkali Scrubbing Test Facility: Summary of
Testing through October 1974. NTIS PB 244-901, June
1975.
Epstein, M. EPA Alkali Scrubbing Test Facility: Advanced
Program. Progress Report prepared by Bechtel for EPA,
September 1976.
Gel'perin, N. I., E. N. Bukharkin, V. Z. Grishko, and M. I.
Tsysin. A Study of contact Heat and Mass Transfer in
Equipment with Fluidized Spherical Packing. Int'l
Chem. Engr., 1^:615-618, 1973.
Goldschmidt, V. W. and M. K. Householder. The Hot Wire
Anemometer as an Aerosol Droplet Size Sampler. Atmos.
Environ., 3_:643-651> 1969.
Johnson, J. M., D. G. Jones, A. Weir, W. C. Martin, and
S. Calvert. Scrubber Experience at Mojave. Paper No.
11. EPA 600/7-76-016, October 1976.
199
-------
Kielback, A. W. The Development of Floating-Bed Scrubbers.
Chem. Engr. Progr. Symposium Series, 5_7_: 51-54, 1961.
Kito, M., M. Sawada, M. Shimada, M. Takata, T. Sakai, and
S. Sugiyama. Gas Holdup in Mobile Beds with Stagnant
Liquid Flow. Kagaku Kogaku Ronbunshu, 2_: 12-15, 1976;
Int'l Chem. Engr., !L6_: 701 - 704 , 1976.
Kito, M., M. Shimada, R. lijima, T. Sakai, M. Takata, and S.
Sugiyama. Liquid-Vapor Interfacial Area for a Liquid
Batch-Type Mobile-Bed Contactor. Kagaku Kogaku Ronbunshu,
2_:16-20, 1976b; Int'l Chem. Engr., 1^:705-709, 1976.
Kito, M., Y. Kayama, T. Sakai, and S. Sugiyama. Minimum
Fluidization Velocity in a Mobile Bed. Kagaku Kogaku
Ronbunshu, 2_: 21-24, 1976c; Int'l Chem. Engr., 16:710-713,
1976c.
Kito, M., T. Monma, Y. Kayama, T. Sagai, and S. Sugiyama.
Pressure Drop and Bed Expansion in a Mobile Bed. Kagaku
Kogaku Ronbunshu, 2_:476-479, 1976d.
Levesh, I. P., N. I. Krainev, and M. I. Niyazov. Calculation
of the Pressure Drop and Heights of Three-Phase Fluidized
Beds. Int'l Chem. Engr., 8^:311-312, 1968.
Levesh, I. P., M. I. Niyazov, N. I. Krainev, and F. F. Ganikhanova,
Mass Transfer in Absorbers with Fluidized Packed Beds.
Int'l Chem. Engr., 8^:379-380, 1968.
McMichael, W. J., J. S. Fan, and C. Y. Wen. Analysis of Sulfur
Dioxide Wet Limestone Scrubbing Data from Pilot Plant
Spray and TCA Scrubbers. Ind. Eng. Chem., Process Des.
Dev., l_5:459-467, 1976.
O'Neill, B. K., D. J. Nicklin, N. J. Morgan, and L. S. Leung.
The Hydrodynamics of Gas-Liquid Contacting in Towers with
Fluidized Packings. Can. J. Chem. Engr., 5_0: 595-601, 1972.
Orr, C. Particulate Technology. The MacMillan Company, New
York, 1966.
Perry, H. Chemical Engineers' Handbook. 5th Ed., McGraw-Hill
Book Company, New York, 1973.
Pollock, W. A., J. P. Tomany, and G. Frieling. Sulfur Dioxide
and Fly Ash Removal from Coal Burning Power Plant. Air
Engr., 24-28, September 1967.
200
-------
Rhudy, R. G. and H. N. Head. Results of Flue Gas Characteri-
zation Testing at the EPA Alkali Wet-Scrubbing Test
Facility. Paper No. 13 in Second EPA Fine Particle
Scrubber Symposium, R. Parker and S. Calvert, eds.
EPA 600/2-77-193, September 1973.
Strumillo, C., J. Adamiec, and T. Kudra. Packed Columns with
Expanding Beds. Int'l Chem. Engr., 14_:652-657, 1974.
Tichy, J., A. Wong, and W. J. M. Douglas. Pressure Drop in a
Mobile-Bed Contactor. Can. J. Chem. Engr., 50:215-220,
1972.
Tichy, J. and W. J. M. Douglas. Bed Expansion in a Mobile-Bed
Contactor. Can. J. Chem. Engr., 5_0: 702-707, 1972.
Tichy, J. and W. J. M. Douglas. Certain Hydrodynamic Char-
acteristics of Mobile-Bed Contactors. Can. J. Chem.
Engr., 5^:618-620, 1973.
Uchida, S., C. S. Chang, and C. Y. Wen. Mechanics of a Tur-
bulent Contact Absorber. Can. J. Chem. Engr., 55:392-396,
1977.
Wen. C., and C. S. Chang. Absorption of S02 in Lime and Lime-
stone Slurry: Pressure Drop Effect on Turbulent Contact
Absorber Performance. Environ. Sci. Tech., 12 : 703- 707,
1978.
Whitmore, P. J. Diffusiophoretic under Turbulent Conditions.
Ph.D. Thesis, University of British Columbia, 1976.
Wozniak, M. and K. Ostergaard. An Investigation of Mass
Transfer in a Countercurrent Three-Phase Fluidized
Bed. Chem. Engr. Sci., 28^167-171, 1973.
Wozniak, M. Pressure Drop and Effective Interfacial Area in
a Column with a Mobile Bed. Int'l Chem. Engr., 17:553-
559, 1977.
201
-------
APPENDIX "A"
DC-1 DROP COUNTER ENTRAINMENT DATA
202
-------
TABLE A-l. DC-1 DROP COUNTER DATA FOR RUN NO. DC-1
Bin
No.
1
2
3
H
5
6
Number
Concen-
tration
(*/cm3)
0.95
1.17
0. 53
0.09
0.07
Cum. Number
Concen-
tration
(»/cm3)
0.95
2.12
2.65
2.74
2.81
Cum.
Entrainment
Volume
(ml/Nm3)
4.00 x 10"6
1.36 x 10'"
1. 76 x 10"3
9.01 X 1(T3
1.69 x ID'3
Drop
Diameter
(um)
1 - 3
3 - 9
9 - 27
27 - 81
81 - 243
>243
TABLE A-2. DC-1 DROP COUNTER DATA FOR RUN NO. DC-2
Bin
No.
1
2
3
4
5
6
Number
Concen-
tration
(*/cm3)
0.43
0.54
0.85
0. 58
0.18
0.27
Cum. Number
Concen-
tration
(#/cm3)
0.43
0.97
1.82
2. 40
2. 58
2.85
Cum.
Entrainment
Volume
(ml/Nm3)
7.6 x 1 0 " '
8.4 x 10'G
1.04 x 10""
6.2 x ID'"
1.92 x 10'3
Drop
Diameter
(ym)
1 - 2
2 - 4
4 - 8
8 - 16
16 - 32
>32
TABLE A-3. DC-1 DROP COUNTER DATA FOR RUN NO. DC-3
Bin
No.
1
2
3
4
5
6
Number
Concen-
tration
(»/cm3)
0.80
0.65
0.92
0.80
0.2-9
0.41
Cum. Number
Concen-
tration
("/cm3)
0.80
1.45
2.37
3.17
3.46
3.87
Cum.
Entrainment
Volume
(ml/Nm3)
1.42 x 10"6
1.06 x ID'5
1.15 x 10""
8.30 x ID'"
2.90 x 10"'
Drop
Diameter
(um)
1 - 2
2 - 4
4 - 8
8 - 16
16 - 32
>32
TABLE A-4. DC-1 DROP COUNTER DATA FOR RUN NO. DC-4
Bin
No.
1
2
3
4
5
6
Number
Concen-
tration
(*/cm3)
0.84
1.35
0.63
0.10
0.09
0.00
Cum. Number
Concen-
tration
(«/cm3)
0.84
2.19
2.82
2.92
3.01
Cum .
Entrainment
Volume
(ml/Nm3)
3.52 x 10'6
1.57 x 10-*
2.08 x 10~3
1 . 0 1) x 1 0 " 2
2.00 x ID'1
Drop
Diameter
(urn)
1 - 3
3 - 9
9 - 27
:~ - si
81 - 243
>243
-------
TABLE A-5. DC-1 DROP COUNTER DATA FOR RUN NO. DC-5
TABLE A-7. DC-1 DROP COUNTER DATA FOR RUN NO. DC-7
ts)
O
Bin
No.
1
2
3
4
5
6
Number
Concen-
tration
(#/cm3)
1.36
2.31
0.92
0.17
0.11
0.00
Cum. Number
Concen-
tration
(#/cm3)
1.36
3.67
4.59
4.76
4.87
Cum.
Entrainment
Volume
(ml/Nm3)
5.70 x 10"6
2.67 x 10'"
3.08 x 10'3
1.70 x 10~2
2.62 x 10"1
Drop
Diameter
(urn)
1 - 3
3 - 9
9 - 27
27 - 81
81 - 243
>243
Bin
No.
1
2
3
4
5
6
Number
Concen-
tration
(#/cm3)
0.94
1.19
1.76
1.19
0.42
0.53
Cum. Number
Concen-
tration
(#/cm3)
0.94
2.13
3.89
5.08
5.50
6.03
Cum.
Entrainment
Volume
(ml/Nm3)
1.66 x 10"6
1.68 x 10~5
2.00 x 10""
1.08 x 10'3
3.04 x 10"3
Drop
Diameter
(urn)
1 - 2
2 - 4
4 - 8
8 - 16
16 - 32
>32
TABLE A-6. DC-1 DROP COUNTER DATA FOR RUN NO. DC-6
TABLE A-8. DC-1 DROP COUNTER DATA FOR RUN NO. DC-8
Bin
No.
1
2
3
4
5
6
Number
Concen-
tration
(#/cm3)
0.82
1.01
1.65
1.06
0.33
0.46
Cum. Number
Concen-
tration
(*/cm3)
0.82
1.83
3.48
4.54
4.87
5.33
Cum.
Entrainment
Volume
(ml/Nm3)
1.45 x 10'6
1.57 x 10'5
2.01 x 10-"
1.16 x 10-3
3.55 x 10'3
Drop
Diameter
(urn)
1 - 2
2 - 4
4 - 8
8 - 16
16 - 32
>32
Bin
No.
1
2
3
4
5
6
Number
Concen-
tration
(S/cm3)
1.69
2.53
1.02
0.17
0.09
0.00
Cum . Numb e r
Concen-
tration
(*/cm3)
1.69
4.22
5.24
5.41
5.49
Cum.
Entrainment
Volume
(ml/Nm3)
7.10 x 10-6
2.93 x 10'*
3.40 x 10'3
1.70 x ID"2
2.18 x 10'1
Drop
Diameter
(ym)
1 - 3
3 - 9
9 - 27
27 - 81
81 - 243
>243
-------
TABLE A-9. DC-1 DROP COUNTER DATA FOR RUN NO.DC-9
TABLE A-ll. DC-1 DROP COUNTER DATA FOR RUN NO. DC-11
IN)
o
On
Bin
No.
1
2
3
4
5
6
Number
Concen-
tration
(#/cm3)
0.98
1.02
1 .81!
1.46
0.54
0.81
Cum. Number
Concen-
tration
(#/cm3)
0.98
2. 00
3.80
5.26
5 . & 0
6.61
Cum.
Entrainment
Volume
(ml/Nm3;)
1.73 x 1C'6
1.61 x 10-5
2.19 x 10'"
1.54 x 10'3
5.45 x 10'3
Drop
Diameter
(urn)
1 - 2
2 - 4
4 - 8
8 - 16
16 - 32
>32
Bin
No.
1
2
3
4
5
6
Number
Concen-
tration
(«/cm3)
1.48
1.12
1.58
0.31
0.20
0. 00
Cum. Number
Concen-
tration
(#/cm3)
1.48
2.60
4.18
4.49
4.69
Cum.
Entrainment
Volume
(ml/Km3)
6.2 x 10'6
1.3 x 10""
5.0 x 10~3
3.1 x 1 0 - 2
4 . S x 1 0 " '
Drop
Diameter
(urn)
1 - 3
3 - 9
9 - 27
27 - 81
81 - 243
>243
TABLE A-10. DC-1 DROP COUNTER DATA FOR RUN NO. DC-10
TABLE A-12 DC-1 DROP COUNTER DATA FOR RUN NO. DC-12
Bin
No.
1
2
3
4
S
6
Number
Concen-
tration
(*/cm3)
1.84
2.35
1.34
0.27
0. 17
II. Illl
Cum. Number
Concen-
tration
(*/cm3)
1.84
4.19
5.53
5. SO
5.97
Cum.
Entrainment
Volume
(ml/Nm')
7.71 x 10~6
2.74 x 10""
4.36 x 10'3
2.60 x 10'z
4.05 x 1 0 ~ l
Drop
Diameter
(ym)
1 - 3
3 - 9
9 - 27
27 - 81
81 - 243
>243
Bin
No.
1
2
3
4
5
6
Number
Concen-
tration
C*/cm3)
0.67
1.13
1.58
1.53
0.47
0.48
Cum. Number
Concen-
tration
(*/cm3)
0.67
1 .80
3.38
4.91
5.38
5.86
Cum.
Entrainment
Volume
(ml/Nm3)
1.19 x 1C'6
1.71 x 10'5
1.96 x 10""
1.58 x 10"3
4.98 x 10"3
Drop
Diameter
(urn)
1 - 2
2 - 4
4 - 8
8 - 16
16 - 32
>32
-------
TABLE A-13. DC-1 DROP COUNTER DATA FOR RUN NO. DC-13
TABLE A-15. DC-1 DROP COUNTER DATA FOR RUN'NO. DC-IS
tx)
O
Bin
No.
1
2
3
4
5
6
Number
Concen-
tration
(»/cm3)
0.26
0.47
0.94
0.72
0.24
0.24
Cum. Number
Concen-
tration
(f/cm3)
0.26
0.73
1.67
2.39
2.63
2.87
Cum.
Entrainment
Volume
(ml/Nms)
4.60 x 10~7
7.10 x 10-6
1.13 x 10""
7.65 x 10'"
2.50 x 10~3
Drop
Diameter
(urn)
1 - 2
2 - 4
4 - 8
8 - 16
16 - 32
>32
Bin
No.
1
2
3
4
5
6
Number
Concen-
tration
(#/cm3)
0.54
1.07
0.42
0.054
0.033
0.00
Cum. Number
Concen-
tration
C#/cm3)
0.54
1.61
2.03
2.08
2.21
Cum.
Entrainment
Volume
(ml/Nm3)
2.27 x 10"6
1.23 x 10""
1.4 x 10"3
5.85 X 10"3
8.00 X 10"2
Drop
Diameter
(pm)
1 - 3
3 - 9
9 - 27
27 - 81
81 - 243
>243
TABLE A-14. DC-1 DROP COUNTER DATA FOR RUN NO. DC-14
TABLE A-16. DC-1 DROP COUNTER DATA FOR RUN NO. DC-16
Bin
No.
1
2
3
4
5
6
Number
Concen-
tration
(#/cm3)
0.69
1.16
0.50
0.08
0.06
0.00
Cum. Number
Concen-
tration
(*/cm3)
0.69
1.85
2.35
2.43
2.50
Cum.
Entrainment
Volume
(ml/Nm3)
2.90 x 10's
1.34 x 10'"
1.66 x 10'3
8.26 x ID'3
1.35 x ID'1
Drop
Diameter
(pm)
1 - 3
3 - 9
9 - 27
27 - 81
81 - 243
>243
Bin
No.
1
2
3
4
S
6
Number
Concen-
tration
(#/cm3)
0.35
0.49
0.66
0.51
0. 14
0.17
Cum. Number
Concen-
tration
(»/cm3)
0.35
0.84
1.50
2.00
2.15
2.32
Cum.
Entrainment
Volume
(ml/Nm3)
.6.20 x 10'7
7.53 x 1C'6
8.21 x 10'5
5.44 x 10'"
1.55 x 10~3
Drop
Diameter
(pm)
1 - 2
2 - 4
4 - 8
8 - 16
16 - 32
>32
-------
TABLE A-17. DC-1 DROP COUNTER DATA FOR RUN NO. DC-17
TABLE A-19. DC-1 DROP COUNTER DATA FOR RUN NO. DC-19
tx)
O
Bin
No.
1
2
3
4
5
6
Number
Concen-
tration
(»/cm3)
0.37
0.30
0.39
0.42
0.08
0.10
Cum. Number
Concen-
tration
(*/cm3)
0.37
0.67
1.06
1.48
1.56
1.66
Cum.
Entrainment
Volume
(ml/Nm3)
6.5 x 10"'
4.85 x ID"6
4.9 x 10'5
4.3 x 10-"
1.0 x 1Q-*
Drop
Diameter
(ym)
1 - 2
2 - 4
4 - 8
8 - 16
16 - 32
>32
Bin
No.
1
2
3
4
5
6
Number
Concen-
tration
(#/cm3)
0.23
0.51
0.20
0.034
0.013
0.00
Cum. Number
Concen-
tration
(I/cm3)
0.23
0.74
0.94
0,97
0.99
Cum.
Entrainment
Volume
(ml/Nm3)
9.6 x 10-'
5.9 x I0-s
6.7 x 10'"
3.5 x 10'3
3.2 x 10-2
Drop
Diameter
(ym)
1 - 3
3 - 9
9 - 27
27 - 81
81 - 243
>243
TABLE A-18. DC-1 DROP COUNTER DATA FOR RUN NO. DC-18
TABLE A-20. DC-1 DROP COUNTER DATA FOR RUN NO. DC-20
Bin
No.
1
2
3
4
5
6
Number
Concen-
tration
(f/cm3)
0.42
0. 74
0. 24
0.035
0.023
0.00
Cum. Number
Concen-
tration
(#/cm3)
0.42
1.16
1.40
1.44
1.46
Cum.
Entrainment
Volume
(ml/Nm!)
1.8 x 10'6
8.6 x 10'5
8.2 x lO'"
3. 7 x ID"3
5.5 x 10"2
Drop
Diameter
(um)
1 - 3
3 - 9
9 - 27
27 - 81
81 - 243
>243
Bin
No.
1
2
3
4
5
6
Number
Concen-
tration
(#/cm3)
0.13
0. 21
0.31
0.25
0.12
0.09
Cum. Number
Concen-
tration
(S/cm3)
0.13
0.34
0.65
0.90
1.02
1 . 10
Cum.
Entrainment
Volume
(ml/Nm3)
2.3 x ID"'
3.2 x 1Q-6
3.8 x 10"5
2.6 x 10'"
1.1 x ID"3
Drop
Diameter
(pm)
1 - 2
2 - 4
4 - 8
8 - 16
16 - 32
>32
-------
TABLE A-21. DC-1 DROP COUNTER DATA FOR RUN NO. DC-21
TABLE A-23. DC-1 DROP COUNTER DATA FOR RUN NO. DC-23
tsj
O
oo
Bin
No.
1
2
3
4
5
6
Number
Concen-
tration
(»/cm3)
0.58
0.44
0.65
0.46
0. 10
0.11
Cum. Number
Concen-
tration
(#/cm3)
1.58
1.02
1.67
2.13
2.23
2.34
Cum.
Entrainment
Volume
(ml /Mm3)
1.0 x ID"6
7.2 x ID'6
8.0 x IQ-S
5.0 x 10-*
1.2 x 10'3
Drop
Diameter
(urn)
1 - 2
2 - 4
4 - 8
8 - 16
16 - 32
>32
Bin
No.
1
2
3
4
5
6
Number
Concen-
tration
(#/cm3)
0.40
0.75
0.42
0.065
0.036
0.00
Cum. Number
Concen-
tration
(*/cm3)
0.40
1.15
1.57
1.64
1.67
Cum.
Entrainment
Volume
(ml/Nm3)
1.7 x ID'6
8.7 x ID'5
1.4 x 10'3
6.8 x 10'3
8.7 x 10'2
Drop
Diameter
(urn)
1 - 3
3 - 9
9 - 27
27 - 81
81 - 243
243
TABLE A-22. DC-1 DROP COUNTER DATA FOR RUN NO. DC-22
TABLE A-24 DC-1 DROP COUNTER DATA FOR RUN NO. DC-24
Bin
No.
1
2
3
4
S
6
Number
Concen-
tration
(»/cm3)
0.66
0.52
0.30
0.045
0.023
0.00
Cum. Number
Concen-
tration
(#/cm3)
0.66
1.18
1.48
1.53
1.55
Cum.
Entrainment
Volume
(ml/Nm3)
2.8 x 10'6
6.2 x 10'5
9.8 x 10-"
4.7 x ID'3
5.6 x 10'2
Drop
Diameter
(vim)
1 - 3
3 - 9
9 - 27
27 - 81
81 - 243
>243
Bin
No.
1
2
3
4
5
6
Number
Concen-
tration
(*/cm3)
0.13
0.25
0.61
0.49
0.16
0.14
Cum. Number
Concen-
tration
(*/cm3)
0.13
0.38
0.99
1.48
1.64
1.78
Cum.
Entrainment
Volume
(ml/Nm3)
2.3 x 10-'
3.7 x 10'6
7.3 x 10-5
5.1 x 10-"
1.7 x ID'3
Drop
Diameter
(urn)
1 - 2
2 - 4
4 - 8
8 - 16
16 - 32
>32
-------
TABLE A-25. DC-1 DROP COUNTER DATA FOR RUN NO. DC-25
Bin
No.
1
2
3
4
5
6
Number
Concen-
tration
(*/cm3)
0. 37
0.52
0.77
0.45
0.16
0.17
Cum. Number
Concen-
tration
(#/cm3)
0.37
0.89
1.66
2.11
2.27
2.44
Cum.
Entrainment
Volume
(ml/Nm3)
6.5 x 10'7
8.0 x 10-6
9.5 x 10-5
5.0 x 10-*
1.7 x 10~3
Drop
Diameter
(Vim)
1 - 2
2 - 4
4 - 8
8 - 16
16 - 32
>32
TABLE A-26. DC-1 DROP COUNTER DATA FOR RUN NO. DC-26
Bin
No.
1
2
3
4
5
6
Number
Concen-
tration
(#/cm3)
0.45
0.91
0.42
0.054
0.03
Cum. Number
Concen-
tration
(#/cm3)
0.45
1.36
1.78
1.83
1.86
Cum.
Entrainment
Volume
(ml/Nm3)
1.9 x 10~6
1.02 x 10-"
1.4 x 1(T3
5.9 x 10'3
7.3 x 10~2
Drop
Diameter
(urn)
1 - 3
3 - 9
9 - 27
27 - 81
81 - 243
>243
TABLE A-27. DC-1 DROP COUNTER DATA FOR RUN'NO. DC-27
Bin
No.
1
2
3
4
5
6
Number
Concen-
tration
(0/cm3)
0.73
1.32
0.51
0.067
0.032
0.00
Cum. Number
Concen-
tration
(#/cm3)
0.73
2.05
2.56
2.63
2.66
Cum.
Entrainment
Volume
(ml/Nm3)
3.10 x 10"6
1. 53 x 10'"
1.71 x 10'3
7.20 x 10'3
7.80 x 10"2
Drop
Diameter
(pm)
1 - 3
3 - 9
9 - 27
27 - 81
81 - 243
>243
TABLE A-28. DC-1 DROP COUNTER DATA FOR RUN NO. DC-28
Bin
No.
1
2
3
4
5
6
Number
Concen-
tration
(#/cm3)
0.53
0.52
1.07
0.57
0.21
0. 19
Cum. Number
Concen-
tration
(«/cm3)
0. 53
1.05
2. 12
2.69
2.90
3. 10
Cum.
Entrainment
Volume
(ml/Nm3)
9.4 x 10~7
8.2 x 10~6
1.3 x 10'"
6.5 x 10-"
2.1 x 10'3
Drop
Diameter
(pm)
1 - 2
? - 4
4 - 8
8 - 16
16 - 32
>32
-------
APPENDIX "B"
TEST CONDITIONS AND PARTICLE DATA
210
-------
TABU; B-I . TEST CONDITIONS AND PARTICLE DATA
MOBILE BED SCRUBBER: 1 - STAGE
PACKING DIAMETER: 3.8 CM
SUPPORTING AND RETAINING GRIDS: HARDWARE SCREEN
AEROSOL: TITANIUM DIOXIDE
Run No.
Cl-1
Cl-2
Cl-3
Cl-4
Cl-S
Cl-6
Cl-7
Cl-8
Cl-9
Cl-10
Cl-11
Cl-12
Cl-13
Cl-14
UG
(»/s)
2.9
2 .9
2.9
2.9
2.9
2.9
2.9
2.9
2.9
2.2
2.2
2.2
2.4
2.4
«G
35.4
35.4
35. 4
35. 4
35.4
35.4
35.4
35.4
35.4
26.9
26.9
26.9
29.5
29.5
"L
293
293
293
293
208
197
197
197
197
227
185
185
269
269
4Pw
5.6
5.6
5.6
5.6
4.6
4.8
5.0
5.2
5.2
S.I
4 .6
4 .6
4.9
4.9
H
s
30
3D
30
30
30
30
30
30
30
3D
30
30
30
30
Hd
(cm)
60
60
60
60
60
63
63
58
63
58
47
47
60
.60
dp.
1.9
2.0
2.3
2.0
2.2
2.3
2.4
2.0
2.3
1.8
2.0
2.6
2.2
2.1
•jmA
1.8
2.1
2.1
2.2
2.1
2.5
2.3
2.0
2.3
2. 0
2.1
2.7
2.3
2. 2
ae
2.3
2.1
1.9
2.0
2.0
2.2
2.0
1.9
2.0
1.7
1.7
1.7
1.9
2.0
2.1
l.D
1 .8
1.8
1.9
1.9
1.9
1.8
1 .8
1.6
1.6
1.6
1.8
1.8
cp, mg/DNn'
37.6
78.6
49.2
71.8
64. .0
143.8
54.3
61 .6
33.6
25.7
24.9
107.9
128.1
31.6
60.2
38.8
58.9
53.2
97.3
44.4
44.1
30.1
21.8
18.4
87.8
103.5
Ft
(I)
84.0
76.6
78.9
82.0
83.1
67.7
81.8
71.6
89 .6
84.5
73.9
81.4
80.8
V
(urnA)
7. 7
8. 6
6.2
7.5
8.0
9.5
7.6
7. 1
8.0
7.0
8.6
9. 0
10.5
TABLE B-2. TEST CONDITIONS AND PARTICLE DATA
MOBILE BED SCRUBBER: 2-STAGE
PACKING DIAMETER: 3.8 CM
SUPPORTING AND RETAINING GRIDS: HARDWARE SCREEN
AEROSOL: RED IRON OXIDE
Run No.
C2-1
C2-2
C2-3
C2-4
C2-5
C2-6
C2-7
C2-8
C2-9
C2-1D
C2-11
C2-12
C2-13
aG
(m/s)
3.1
3.1
2.6
2.6
2.6
2. 6
2.0
2.0
2.0
2.0
2.0
2.0
3.1
«G
(m3/min]
37.9
37.9
31.2
31.2
31.2
31.2
24.0
24.0
24.0
24.0
24.0
24.0
37.9
«L
(i/»in)
136
136
136
136
136
136
136
227
227
310
136
136
136
4P«
5.3
4.9
5. 0
5.0
5.0
4 . 7
4.8
5.7
5.7
6.0
8.9
8.9
9.4
Hs
15
IS
15
15
15
15
15
15
15
15
30
30
30
Hd
tern)
32
32
30
30
30
32
30
30
30
30
43
43
60
V
1.8
1.7
1.9
2. 2
1.9
1.9
2.1
1.9
2.3
2.4
1.9
1.9
2. 4
umA
1.9
1.7
1.9
2.1
2.0
1 .8
2.1
1.8
2.3
2.1
1.8
1.8
2.4
a
1.9
1.8
1.9
1.9
2.0
1 .7
1.9
1.9
2.0
2.0
2.2
2.1
2.0
E
1.8
1.7
1.8
1.8
1.9
1.7
1.8
1.9
1.9
2.0
1.8
1.7
1.8
v-
52.0
66.3
57. S
60.5
75.9
36.2
69.1
110.4
97.4
71.7
49.3
57.1
165.3
g/DN«s
36.1
53.1
38.0
41.5
58.6
31.1
49.1
75.1
73.2
47. S
28.1
31.7
94.9
Ft
(t)
69.4
80.1
66.1
68 .6
72.2
85.9
71.1
68.0
75.2
66.2
57.0
55. S
57.4
V
(umA)
7.9
6.8
7.4
9.2
9. 8
7 .6
9.6
8.5
9.0
6.0
3.6
3.5
5.5
TEST CONDITIONS AND PARTICLE DATA
MOBILE BED SCRUBBER: 2-STAGE
PACKING DIAMETER: 2.S CM
RETAINING AND SUPPORTING GRIDS: HARDWARE SCREENS
AEROSOL: CUPOLA DUST
Run No.
C2-14
C2-15
C2-16
C2-17
C2-18
C2-19
C2-20
C2-21
"G
(m/s)
2.9
2.9
2.9
2.9
4.1
4.1
4.1
4.1
"G
(mVmin)
35.1
35. 1
35.1
35.]
49.6
49.6
49.6
49.6
«L
U/min)
269
269
352
352
269
269
352
307
1P»
(cm B.C.)
10.5
10. 2
13.6
13.5
10.3
10.0
14.3
12.6
KS
(cm)
23
23
23
23
23
23
23
23
Hd
(cm)
S3
53
60
60
70
70
72
72
dpP
Inlet
2.4
1.9
3.4
2.5
2.8
2.6
3.1
2.9
umA
Outlet
1.9
1.8
1.6
1.9
1. 7
1.6
1.4
1.4
"*
Inlet
2 . 3
2.5
3.1
2.2
3.0
2.7
2.9
2.7
Outlet
1.8
1.9
1.9
2.0
1.9
1.9
1.6
1.9
cp, mg/DNm'
Inlet
57.3
77.8
86.2
49.1
52.1
64.6
79.7
98.1
Outlet
29.9
33.2
37.6
21.6
22.4
26.8
11.3
26.4
Ft
(*)
52.2
42. 7
43.6
44.0
43. 0
72.4
14.2
26.9
V
(umA)
2.9
2.9
3.0
2. 7
3.3
3.1
1.8
1.7
211
-------
TEST CONDITIONS AND PARTICLE DATA
MOBILE BED: 3-STAGE
PACKING DIAMETER: 3.8 CM
SUPPORTING AND RETAINING GRIDS: HARDWARE SCREEN
AEROSOL: RED IRON OXIDE
ts)
h-i
to
Run No.
C3-1
C3-2
C3-3
C3-4
C3-5
C3-6
C3-7
C3-8
C3-9
C3-10
C3-11
C3-12
C3-13
C3-14
C3-15
C3-16
C3-17
C3-18
C3-19
C3-20
C3-21
C3-22
C3-23
C3-24
C3-25
C3-26
C3-27
C3-28
C3-29
C3-30
C3-31
C3-32
C3-33
C3-34
C3-35
C3-36
C3-37
C3-38
UG
Cm/s)
2.6
2.6
2.6
2.6
3.1
3.1
3.1
3.1
3.1
3.1
3.1
3.1
3.1
3.1
2.8
2.8
2.8
2.8
2.8
2.0
2.0
2.0
2.0
2.0
2.0
2.3
2.3
2.3
2.3
2.3
2.3
2.3
2.3
2.3
2.3
2.3
2.3
2.6
«G
(ra'/minj
31.2
31.2
31.2
31.2
37.9
37.9
37.9
37.9
37.9
37.9
37.9
37.9
37.9
37.9
34.3
34.3
34.3
34.3
34.3
24.0
24.0
24.0
24.0
24.0
24.0
27.8
27.8
27.8
27.8
27.8
27.8
27.8
27.8
27.8
27.8
27.8
27.8
31.1
<»L
(J/rain)
227
227
310
310
310
310
310
227
227
227
136
136
136
136
136
136
227
227
310
136
136
227
227
310
310
136
136
227
227
310
310
136
136
227
227
310
310
136
APw
(cmW.C.)
8.4
8.4
10.4
10.4
10.9
10.9
10.9
•8.9
8.9
8.9
7.7
7.7
7.7
7.7
8.3
8.3
9.2
9.2
11.2
7.3
7. 3
7.5
7.5
9.0
9.0
7.5
7.5
8.9
8.9
10.0
10.0
9.8
9.8
10.9
10.9
13.1
13. 1
11.1
H
(cm)
15
15
15
15
15
15
15
15
IS
IS
15
IS
IS
15
15
IS
15
15
15
15
15
15
15
15
15
15
IS
15
IS
15
15
23
23
23
23
23
23
23
"d
(cm)
32
32
38
38
44
44
44
38
38
38
35
35
35
38
35
35
38
38
43
28
28
28
28
30
30
32
32
32
32
32
32
40
40
43
43
47
47
43
dF8
Inlet
2.2
2.1
2.1
2.1
2".l
2.0
2.3
2.1
2.1
2.2
2.0
2.2
2.7
2.4
2.3
2.2
2.6
2.1
2.8
2.4
2.3
2.3
2.6
2.3
2.2
2.4
2.4
2.3
2.1
2.1
2.1
2.2
2.1
2.2
2.1
2.1
2.2
2.2
ymA
Outlet
2.0
1.9
1.9
1.9
2.0
2.0
2.2
1.9
1.8
2.0
2.0
2.4
2.5
2.3
2.2
2.1
2.3
2.0
2.1
2.2
2.2
2.2
2.1
2.1
2.0
2.2
2.1
2.2
2.0
2.0
1.9
2.0
1.9
1.8
2.0
1.9
2.0
1.9
°e
Inlet
2.0
1.9
1.9
2.0
2.0
1.8
2.1
2.1
2.0
1.8
2.0
1.9
1.9
1.8
1.7
1.9
2.0
2.0
2.1
1.8
1.9
1.8
1.9
1.9
1.8
1.7
1.9
1.8
1.6
1.8
1.9
1.7
1.9
1.9
1.8
1.7
1.8
1.9
Outlet
1.9
1.7
1.8
1.8
1.7
1.6
1.7
1.7
1.7
1.7
1.7
1.8
1.8
1.7
1.7
1.7
1.8
1.7
2.0
1.7
1.8
1.7
1.8
1.8
1.7
1.7
1.7
1.7
1.6
1.6
1.8
1.6
1.7
1.6
1.6
1.7
1.6
1.7
c_, mg/DNm!
Inlet
66.7
74.4
58.7
43.9
74.3
46.5
41.7
50.9
53.0
51.3
63.5
47.8
81.1
72.2
99.3
51.5
71.3
51.5
102.0
104.5
122.7
88.4
81.0
84.2
96.7
110.7
127.0
95.0
128.1
134.6
92.7
120.4
11S.1
119.6
138.9
132.0
122.7
91.3
Outlet
47.8
51.6
37.8
34.0
45.3
29.4
23.8
31.7
36.7
35. 5
48.7
38.5
53.7
49.3
65.5
33.9
43.1
33.0
SO. 8
75.3
82.1
56.6
49.5
55.9
46.6
73.7
79.0
58.7
79.1
70.9
67.6
86.9
82.4
83.4
87.1
78.0
67.1
54.6
Ft
w
71.7
69.4
64.4
77.4
61.0
63.2
57.1
62.3
69.2
69.2
76.7
80.5
66.2
68.3
66.0
65.8
60.4
64.1
49.8
72. -1
66.9
64.0
61.1
66.4
48.2
66.6
62.2
74.2
61.7
52.7
72.9
72.2
71.6
69.7
62.7
59.1
54.7
59.8
d
PC
(jjmA)
5.2
4.2
4.3
4.5
4.1
3.2
3.S
4.4
3.8
4.6
5.2
5.7
4.3
5.3
5.0
4.5
4.6
3.7
2.7
5.3
6.0
5.0
3.8
4.2
2.5
5.2
4. 1
4.5
4.1
3.1
4.5
4.1
4.2
3.4
3.4
2.9
2.8
3.4
-------
TABLE B-5. TEST CONDITIONS AND PARTICLE DATA
MOBILE BED SCRUBBER: 3-STAGE
PACKING DIAMETER: 1.6 CM
SUPPORTING AND RETAINING GRIDS: HARDWARE SCREEN
AEROSOL: FLY ASH
Run No.
C3-39
C3-40
C3-41
C3-42
C3-43
C3-44
C3-45
C3-46
UG
(m/s)
2.3
2.3
2.3
2.3
2.9
2.9
2.9
2.9
"G
27. B
27.8
27.8
27.8
35.1
35.1
35.1
35.1
IL
(*/min)
310
310
136
136
310
310
227
227
4P»
13.1
13.1
9.6
9.6
13.8
13.8
11.6
11.6
"s
23
23
23
23
23
23
23
23
Hd
47
47
35
43
53
53
46
46
V
2.4
2.7
2.1
2.3
3.7
2.4
2.5
2.2
jjmA
1.8
2.0
2.0
1.7
2.0
1.9
2.1
2.1
OE
Inlet
2.3
3.4
2.4
2.8
2.8
2.4
2.2
2.4
1.6
2.2
2.2
2.0
2.3
2.5
2.1
2.2
CD, mg/DNm'
28.7
31.4
21.4
27.7
19.9
48.1
73.6
45.4
9.7
12.0
12.9
12.6
9.7
29.0
33.2
24.0
PT
(*)
33.8
38.2
60.3
45.5
48.7
60.3
45.1
53.2
V
(pmA)
2.4
3.2
S.2
3.0
2.4
2.4
2.8
3.1
TABLE B-6. TEST CONDITIONS AND PARTICLE DATA
MOBILE BED SCRUBBER: 3-STAGE
PACKING DIAMETER: 3.8 CM
SUPPORTING AMD RETAINING GRIDS: HARDWARE SCREEN
AEROSOL: CUPOLA DUST
Run No.
C3-47
CJ-48
C3-49
C3-50
C3-51
C3-52
C3-53
C3-54
C3-55
C3-56
C3-57
C3-58
C3-59
C3-60
C3-61
C3-62
C3-63
C3-64
C3-65
C3-66
C3-67
C3-68
C3-69
C3-70
C3-71
C3-72
C3-73
UG
(m/s)
2.9
2.9
2.9
2.9
2.9
2.3
2.3
3.9
3.9
3.9
3.9
3.9
3.9
3.9
3.9
3.9
3.9
3.9
4.4
4.4
4.1
4.1
4.1
4.1
4.1
4.1
4.1
%
(.'/"in)
35.1
35.1
35.1
35.1
35.1
27.8
27.8
47.0
47.0
47.0
47.0
47.0
47.0
47.0
47.0
47.0
47.0
47.0
53.8
53.8
50.0
50.0
50.0
50.0
50.0
50.0
50.0
«L
136
136
227
310
310
310
310
227
227
227
227
310
310
136
352
352
401
401
136
136
136
136
227
227
310
310
310
iP»
10.2
10.2
11.6
14.2
13.8
13.6
13.6
12.5
12.5
12.5
12.5
'14.3
14.3
9.7
15.4
15.4
17.8
17.1
9.9
10.6
9.7
9.5
12.4
11.9
14.4
14.6
13.8
«s
23
23
23
23
23
23
23
23
23
23
23
23
23
2!
23
23
23
23
23
23
23
23
23
23
23
23
25
«d
43
43
46
46
46
47
47
53
53
53
53
62
62
50
68
68
70
70
65
65
58
58
62
62
68
68
68
"PR
2.2
2.3
2.4
2'. 6
2.0
2.3
3.8
5.0
2.8
2.8
2.8
2.5
2.2
2.1
2.0
2.6
2.5'
2.7
2.2
2.9
2.6
2.4
2.5
3.2
3.2
3.0
11 mA
2.0
2.1
1.8
1.8
1.5
1.7
2.1
1.8
2.1
1.6
1.5
1.3
1.8
1.5
1.5
1.6
1.3
1.9
1.7
2.2
1.8
1.7
2.0
1.9
1.8
1.5
°«
2.2
2.4
2.5
2.4
2.4
2.5
2.6
2.6
2.2
2.9
2.5
2.6
2.8
2.4
2.5
2.1
3.1
2.0
2.6
2.9
3.3
2.7
2.5
2.6
2.6
3.0
1.8
2.0
l.B
1.9
1.9
1.9
1.9
2.2
1.9
1.7
2.1
1.7
1.8
1.7
2.3
1.7
1.7
2.0
2.0
2.3
2.1
1.8
2.1
1.8
1.7
1.8
cp, ng/DNm'
Inlet
30.8
65.3
63.3
38.0
47.8
90.3
86.1
14.6
34.9
57.3
23.3
40.1
24.7
64.3
82.0
65.9
60.9
41.4
47.2
100.0
90.8
85.8
96.1
110.0
59.0
72.1
14.7
28.6
25. 8
15.9
17.5
35.6
32.1
4.0
17.5
18.5
11.3
16.4
14.8
27.1
40.3
29.2
23.7
27.3
28.8
49.1
41.3
33.9
38.1
28.7
16.9
28.0
Ft
(1)
47.7
43.8
40.8
41.8
36.6
39.4
37.3
27.4
50.1
32.3
48.5
40.9
59.9
42.1
49.1
44.3
36.9
65.9
61.0
49.1
45.5
39.5
39.6
26.1
28.6
36.8
Jpc
(umA)
3.1
2.9
2.7
2.4
1.6
2.2
2.2
2.2
2.5
2.3
2.4
2.1
3.9
2.2
2.1
2.3
2.2
3.3
3.6
4.0
3.4
2.7
2.2
2.0
1.9 ,
2.5
213
-------
TABLE B-7. TEST CONDITIONS AND PARTICLE DATA
MOBILE BED SCRUBBER: 3-STAGE
PACKING DIAMETER: 5.1 CM
RETAINING AND SUPPORTING GRIDS: HARDWARE SCREEN
AEROSOL: CUPOLA DUST
Run No.
C3-74
C3-75
C3-76
C3-77
C3-78
C3-79
C3-80
C3-81
UG
(m/s)
2.9
2.9
2.9
2.9
4.1
4.1
4.1
4.1
QG
(m3/min)
35.1
35.1
35.1
35.1
50.0
50.0
50.0
50.0
QL
Cl/min)
227
269
352
352
269
269
352
352
APw
ConW.C.)
9.7
10.1
11.8
11.8
10.5
10.8
13.1
13.2
Hs
Con)
23
23
23
23
23
23
23
23
Hd
(cm)
43
43
53
53-
65
65
68
68
d
PK
Inlet
2.4
2.5
2.1
2.1
2.1
2.5
2.6
2.5
ymA
Outlet
1.6
1.7
1.6
1.8
1.8
2.9
1.5
1.6
°*
Inlet
2.8
2.8
2.5
2.3
2.6
1.7.
2.6
2.9
Outlet
2.2
2.1
2.2
1.9
1.9
1.9
1.8
2.1
cp, rag/mm'
Inlet
114.0
114.0
122.0
136.1
69.8
69.0
80.3
64.8
Outlet
59. S
56.6
58.7
54.6
38.4
31.4
35.7
32.4
PT
m
52.2
49.6
48.1
40.1
55.0
45.5
44.5
50.0
V
OmA)
3.1
3.3
2.9
2. 2
3.2
3.1
2.5
2.7
to
TABLE B-8. TEST CONDITIONS AND PARTICLE DATA
MOBILE BED: 3-STAGE
PACKING DIAMETER: 3.B CM
RETAINING AND SUPPORTING GRIDS: PLASTIC NETS
AEROSOL: CUPOLA DUST
Run No.
C3-82
C3-83
C3-84
C3-8S
C3-86
C3-87
C3-88
C3-89
C3-90
C3-91
C3-92
C3-93
C3-94
C3-95
C3-96
C3-97
C3-98
C3-99
UG
(m/s)
4.2
4.2
4.2
4.2
4.2
4.2
4.2
4.1
4.1
4.2
4.2
2.1
2.1
2.1
2.1
2.1
3.0
3.5
"G
(mVrain)
50.4
50.4
50.4
50.4
50.4
50.4
50.4
49.6
49.6
50.4
50.4
25.5
25.5
25.5
25.5
25.5
36.3
42.6
\
(I/rain)
363
322
322
322
322
280
280
280
280
237
237
237
237
280
280
322
387
269
4Pw
(cmW.C.)
39.5
31.2
34.2
33.0
33.0
21.3
23.4
25.9
25.4
18.3
18.3
14.5
13.1
15.1
15.5
16.8
25.4
18.3
"s
(cm)
23
23
23
23
23
23
23
23
23
23
23
23
23
23
23
23
23
23
Hd
(cm)
70
70
70
70
70
70
70
70
70
66
66
43
43
47
50
55
V
Inlet
2.S
4.8
2.3
2.9
3.2
2.0
2.6
2. 5
2.2
2.3
1.4
1.6
1.2
1.1
1.0
1.1
2.2
2.0
vi mA
Outlet
0.96
1.2
1.1
1.0
1.1
1.2
1.3
1. 1
0.93
1.4
1.2
1.3
1.2
1.2
1.2
1.3
1.2
1.5
°g
Inlet
2.4
3.1
2.4
2.5
2.5
2.6
2.5
2.8
2.2
2.6
2.7
2.7
2.4
2.8
2.9
3.3
2.7
2.5
Outlet
1.6
1.4
1.7
1.8
1.8
2.0
1.8
1.9
1.6
1.9
1.7
1.7
1.6
1.6
1.6
1.4
1.7
1.9
c , mg/DNm3
Inlet
92.3
150.3
93.0
105.0
97.0
128.0
90.3
96.2
100.3
106.0
104.0
160.0
114.0
120.0
137.0
190.0
67.2
52.9
Outlet
13.0
8.2
9.8
16.4
14.4
38.6
25.3
12.3
19.5
41.3
24.9
93.0
52.6
50.7
50.3
63.2
13.0
18.2
Pt
(*)
14.1
5.5
10.5
15.6
14.8
30.2
28.0
12.8
19.4
39.0
23. .7
58.1
46.1
42.3
36.7
33.3
19.3
34.5
V
(ymA)
0.93
*
*
0.96
1.1
1.1
1.5
ft
1.0
2.0
*
2.7
2.2
2.2
2.1
2.1
0.96
1.7
-------
TABLE B-9. SLURRY SCRUBBING TEST CONDITIONS AND PARTICLE DATA
MOBILE BED SCRUBBER: 3-STAGE
PACKING DIMETER: 3.8 cm
SUPPORTING AND RETAINING GRID: HARDWARE SCREEN
AEROSOL: CUPOLA DUST
ISJ
Run No.
S3-1
S3-2
S3-3
S3-4
S3-S
S3-6
S3-7
S3-8
S3-9
S3-10
S3-11
S3-12
S3-13
S3-14
S3-15
S3-16
S3-17
UG
(m/s)
2.8
2.8
3.4
3.4
2. 8
2.8
2.8
3.4
3.4
2.8
2.8
3.4
3.4
2.8
2.8
3.4
3.4
QG
(mVmin)
34
34
41
41
34
34
34
41
41
34
34
41
41
34
34
41
41
QL
(Jl/min)
227
310
227
310
227
227
310
227
310
227
310
227
310
227
310
227
310
APw
(cmW.C.)
12.0
14.1
12.3
14.3
12.0
12.0
13.8
12.3
14.1
12.0
13.5
12.6
14.0
11.7
12.9
12.2
13.8
H
(cm)
23
23
23
23
23
2-3
23
23
23
23
23
23
23
23
23
23
23
wt &
in
Slurry
5
S
5
5
10
10
10
10
10
10
10
10
10
10
10
10
10
V
Inlet
2.4
2.4
2.5
2.5
2.5
2.6
3.0
3.0
2.3
--
--
--
--
--
--
--
--
ymA
Outlet
2.0
1.6
1.9
1.7
1.7
1.8
1.8
1.9
1. 7
--
--
--
--
--
--
--
--
°*
Inlet
2.1
2.3
2.2
2.2
2.2
2. 3
2.3
2.6
2.1
--
--
--
--
--
--
--
--
Outlet
1.9
1.9
2.1
1.8
1.9
1.8
2.0
1.9
1.8
--
--
--
--
--
--
Cp, mg/DNra3
Inlet
38.8
37.9
56.4
45.5
76.3
83.0
87.1
73.9
73.4
--
--
--
--
--
Outlet
16.6
13.5
29.8
21.9
37.1
34.5
30.9
28.6
30.7
0.2
0. 2
0.3
0.3
0.2
0.2
0.3
0.4
Ft
(*)
42.8
35.6
52.8
48.1
48.6
41.6
35.5
38.7
41.8
--
--
--
V
(pmAj
2.3
1.9
3.0
2.6
2. 5
2.7
2.2
2.7
1.9
--
-
--
-------
TABU; n-io •
r-o
F/C SCRUBBING TEST CONDITIONS AND PARTICLE DATA
MOBILE BRP: 3-STAGE
tth = 3.8 CM
Hs • 23 CM
RETAINING AND SUPPORTING GRIDS: HARDWARE SCREEN
AEROSOL: CUPOLA DUST
Run
No
FC-1
FC-2
FC-3
FC-4
FC-5
FC-6
FC-7
FC-8
FC-9
FC-10
FC-11
FC-I2
FC-13
FC-14
FC-15
FC-16
FC-17
FC-1 8
FC-19
FC-20
FC-21
FC-22
UG
(m/s)
2.8
3.1
3.4
3.1
3.1
3.1
3.1
3.1
3.1
3.1
3.1
3.1
3.1
2.8
2.8
2.8
3.1
3.4
3.4
3.1
3.4
3.4
"G
(m Vmin)
34
38
41
38
38
38
38
38
38
38
38
38
38
34
34
34
38
41
•11
38
41
41
-------
APPENDIX "C"
GRADE PENETRATION CURVES FOR
COLD OPERATION MODE
217
-------
SINGLE STAGE MOBILE BF.D
HARDWARE SCREEN SUPPORT
UG 2!)() cm/s
QL/Q(, 8.3 l/m1
APW= 5.6 cm W.C.
db 3.8 cm
HS 30 cm
TITANIUM DIOXIDE AEROSOL
I
1 t
RUN NO. Cl-S
SINGLE STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
UG 290 cm/s
0../Qr 5.9 t/m1
APW 5.0 cm W.C.
d, 3 . 8 cm
b
II 30 cm
TITANIUM DIOXIDE AEROSOL
AERODYNAMIC PARTICLE DIAMETER,
1 5 10
AERODYNAMIC PARTICLE DIAMETER, umA
Figure C-L. Experimental grade penetration
curves.
Figure C-2.
Experimental grade penetration
curve.
0.05
. SINGLE STAGE MOBILE BED
u^ 290 cra/s
QL/QG 5.6 l/m'
&PW 5.1 cm W.C.
d-. 3.8 cm
b
H 30 cm
TITANIUM DIOXIDE AEROSOL
I
I
_j I
5 10 50
AERODYNAMIC PARTICLE DIAMETER, umA
RUN NO. Cl-10
~ SINGLE STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
UG 220 cm/s
QL/QG 8.4
-------
1.0
0.5
RUN NO. Cl-11
Cl-12
SINGLE STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
Ug 220 cm/s
QL/Qfj - 6.9 l/m*
iP,,= 4.6 cm W.C.
H
3.8 cm
30 cm
TITANIUM DIOXIDE AEROSOL
i i i I l i i i i
50
AERODYNAMIC PARTICLE DIAMETER,
1.0
0.1
0.01
RUN NO. Cl-13
Cl-14
SINGLE STARE MOBILE BED
HARDWARE SCREEN SUPPORT
UG 2JO cm/s
QL/QG' 9.1 «./m3
APW = 4.9 cm W.C.
d, 3.8 cm
D
HS 30 cm
TITANIUM DIOXIDE AEROSOL
10
AERODYNAMIC PARTICLE DIAMETER, umA
50
Figure C-5. Experimental grade penetration
curves.
Figure C-6. Experimental grade penetration
curves.
0.1
0.05
0.01
RUN NO. C2-1
C2-2
2-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
UG • 310 cm/s
APW = 5.1 cm W.C.
d. 3.8 cm
D
HS = 15 cm
RED IRON OXIDE AEROSOL
I i i i i i i I I
1.0
0.01
2-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
QL/QG 4.4 l/m>
APW- 5.0 cm W.C.
d. 3.8 cm
D
HS = 15 cm
RED IRON OXIDE AEROSOL
1 I I I I I I I I
I
t
AERODYNAMIC PARTICLE DIAMETER, umA
1 5 10
AERODYNAMIC PARTICLE DIAMETER, umA
Figure C-7. Experimental grade penetration
curves.
Figure C-8. Experimental grade penetration
curves.
219
-------
t 0.05
RUN NO. C2-7
2-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
UG = 200 cm/s
QL/Qg 5.7 l/m!
4P..- 4.7 cm W.C.
Hs 15 cm
RED IRON OXIDE AEROSOL
1 5 10 50
AERODYNAMIC PARTICLE DIAMETER, umA
1 . 0
0.5
c
o
u
(-
4-i
2
O
f-
3 °'1
E-H
w
2:
OJ
PL,
w 0.05
u
E-
K
<:
&,
0.01
1 —
: '
- RUN NO.
_
~
-
-
_ 2-STAGE
HARDWARE
uf =
. VG =
AP =
- db '
Hs -
1 1 1 1 1 1 1 1
——-^
^^>^
C2-8-^ ^^-"^\
^ —
C2-9 '
MOBILE BED
SCREEN SUPPORT
200 cm/s
9.5 «./mB
5.7 cm W.C.
3.8 cm
15 cm
1 1 ~
~
*^~^-
"s.
\
"
-
-
_
_
_
_
RED IRON OXIDE AEROSOL
1
i l i 1 i i i l
1 1 1
1 5 10
AERODYNAMIC PARTICLE DIAMETER, umA
SO
Figure C-9, Experimental grade penetration
curve.
Figure C-10. Experimental grade penetration
curves.
1.0
. RUN NO. C2-10
2-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
UG 200 cm/s
QL/QG 12.9 i/m3
4PW^ 6 cm W.C.
dfc 3.8 cm
HS = 15 cm
RED IRON OXIDE AEROSOL
I t I I 1 1 I .1 I
5 10 50
AERODYNAMIC PARTICLE DIAMETER,'ymA
1.0
0.5
0.05
0.01
I III
RUN NO. C2-11
C2-12
2-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
APK- 9.9 cm W.C.
"G 200 cm/s
5.7 H/m'
3.8 cm
HS 30 cm
RED IRON OXIDE AEROSOL
V^C
dv
5 10 50
AERODYNAMIC PARTICLE DIAMETER, pmA
Figure C-ll. Experimental grade penetration
curve.
Figure C-12. Experimental '.;rade penetration
curves.
220
-------
1.0
0.5
0.05
0.01
RUN NO. C2-13
2-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
UG * 310 cm/s
AP,.,= 10.4 cm W.C.
3.8 cm
HS 30 cm
RED IRON OXIDE AEROSOL
1 S 10
AERODYNAMIC APRTICLE DIAMETER,
0.1
0.05
0.01
RUN NO. C2-14
C 2 -1 5
2-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
ufi • 290 cm/s
qL/Q-. - 7.7 4/m!
APW 10.4 crn W.C.
dfa 2.5 cm
H, 23 cm
CUPOLA DUST AEROSOL
0.2 0.5 1 5 10
AERODYNAMIC PARTICLE DIAMETER, umA
Figure C-13. Experimental grade penetration
curve.
Figure C-14. Experimental uradc penetration
curves.
0.5
0.1
0.05
0.01
RUN NO. C2-16
C2-17
2-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
UG
290 cm/s
10.0 H/m3
13.6 cm W.C.
HS 23 cm
CUPOLA DUST AEROSOL
0.2 0.5 1 5 10
AERODYNAMIC PARTICLE DIAMETER, ymA
0.05
0.01
RUN NO. C2-18
C2-19
2-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
UQ 410 cm/s
Q /Q 5.4 l/n'
AP - 10.2 cm W.C.
H 23 cm
CUPOLA DUST AEROSOL
i i I i
0.2 0.5 1 5 10
AERODYNAMIC PARTICLE DIAMETER, pmA
Figure C-15. Experimental grade penetration
curves.
Figure C-16. Experimental
curves .
penetration
221
-------
0.5
RUN NO. C2-20
• 2-STAGE MOBILE BED
• HARDWARE SCREEN SUPPORT
ufi = 410 cm/s
Q /Q =7.1 i/m'
APW - 14.3 cm K.C.
d. 2.5 cm
HS 23 cm
CUPOLA DUST AEROSOL
0.2 0.5 1 5 10
AERODYNAMIC PARTICLE DIAMETER, umA
0.05
RUN NO. C2-21
2-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
UG 410 cm/s
Qj/Qc 6.2 It/in'
4PW 12.6 cm W.C.
dfc 2.5 cm
HS 23 cm
CUPOLA DUST AEROSOL
0.2 0.5 1 5 10
AERODYNAMIC PARTICLE DIAMETER, ymA
Figure C-17. Experimental grade penetration
curve.
Figure C-17a. Experimental grade penetration
curve.
0.1
0.05
RUN NO. C3-1
C3-2
3-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
UG 260 cm/s
QL/QG 7.3 i/m'
AP 8.4 cm W.C.
dfe 3.8 cm
HS 15 en
RED IRON OXIDE AEROSOL
1.0
1 5 10
AERODYNAMIC PARTICLE DIAMETER,
RUN NO. C3-3
C3-4
3-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
"G
QL/QG
260 cm/s
9.9 £/m3
4P - 10.4 cm W.C.
dfc 3.8 cm
HS 15 cm
RED IRON OXICE AEROSOL
AERODYNAMIC PARTICLE DIAMETER, umA
I-'igure C-18. Experimental grade penetration
curves.
Figure C-19. Experimental grade penetration
curves.
222
-------
1 . U
0. 5
0.1
0.05
0.01
0.
- ^v ;
RUN NO. ci-5~~~~^\^\\i
C3-6 ""^-^-"XVi
\Y
.
'. 3 -STAGE MOBILE BED
. HARDWARE SCREEN SUPPORT
u_ 310 cm/s
b
Qi/QG 8- 2 l/m!
APW 10.9 cm W.C.
d, = 3.8 cm
b
HS = 15 cm
RED IRON OXIDE AEROSOL
i i i i 1 i i i i i i i i i i i
2 0.5 1 5 10
1.0
0.1
0.05
0.01
3-STAnn MOBILE BED
- HARDWARE SCREEN SUPPORT
ufi 310 cm/s
APW 8.9 cm W.C.
d, = 3.8 cm
D
HS = 15 cm
RED IRON OXIDE AEROSOL
i
i i i
AERODYNAMIC PARTICLE DIAMETER, uraA
.2 0.5 1 5 10
AERODYNAMIC PARTICLE DIAMETER, |;nA
Figure C-20. Experimental urailc penetration
curves.
Figure C-21. Experimental .^riidc pcncr ra r i mi
curves.
0.5
0.05
: i^. :
: y/vV -
RUN NO. C3-11''//\X::^\\V •
C3-12 *jS ^-"^Y\
/ ^^^ \ y\ "
C3"14.^^"^ \ ' -
.
I ;
. 3-STAGE MOBILE BED I
- HARDWARE SCREEN SUPPORT
UG = 310 cm/s
' QL/0G - 3.6 t/m!
APW = 7.7 cm W.C.
d. = 3.8 ce
b
H - 15 cm
s
RED IRON OXIDE AEROSOL
X . U
0.5
c
o
4-1
u
B
(-
"4-1
z
o
H
$ 0.1
&
if.
UJ
0.
tu 0.05
^j
u
i— i
H
cc:
<
a.
0.01
: ' ' 1 ' ' " 1 ^" ' ' I ' ' "-
1 J^\^ ~
RUN NO. C3-15 J^/ ^s. .
C3-16 ^ \\
\
\ *
\
— —
"
- 3-STAGE MOBILE BED
- HARDWARE SCREEN SUPPORT
~~ UG 280 cm/s ~~
- QL/QG 4 t/ms
APw 8.3 cm W.C.
d, 3.8 cm
b
H IS cm
s
RED IRON OXIDE AEROSOL
. . 1 . , , ,1 , .,!.,,,
0.2 0.5 1 5
AERODYNAMIC PARTICLE DIAMETER, umA
Figure C-22. Experimental grade penetration
curves.
AERODYNAMIC PARTICLE DIAMETER, umA
Figure C-23. Experimental «r;ulc pcnet rat i on
curves.
223
-------
0.5
0.1
m O.OS
T— I
I | I I
RUN NO. C3-17
C3-18
• 3-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
UG = 280 cm/s
QL/QG 6.6 Jl/m3
iPw - 9.2 cm W.C.
dfc - 3.8 cm
HS - 15 cm
RED IRON OXIDE AEROSOL
I , I , I , ,1
0.2
0.5
1
AERODYNAMIC PARTICLE DIAMETER, umA
1.0
0.5 -
w 0.05
0.01
RUN NO. C3-19
3-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
u
'G
280 cm/s
9 i/m'
AP^ - 11.2 cm W.C.
db 3.8 cm
HS 15 cm
RED IRON OXIDE AEROSOL
, . I , . , .1
0.2 0.5 1 5
AERODYNAMIC PARTICLE DIAMETER, jim
Figure C-24. Experimental grade penetration
curves.
Figure C-25. Experimental grade ncnetration
curve.
1.0
0.05
RUN NO. C3-20
C3-Z1
3-STAGE MOBILE BED
- HARDWARE SCREEN SUPPORT
V<»G
AP
200 cm/s
5.7 1/m1
7.3 cm W.C.
d. = 3.8 cm
b
S 15 cm
RED IRON OXIDE AEROSOL
I
,1
J_
0.2 0.5 1 5 10
AERODYNAMIC PARTICLE DIAMETER, umA
Figure C-26. Experimental 'jrude [icnct rnt i on
curves.
0.5
0.1
O.OS
0.01
RUN NO. C3-22
C3-23
QL/QQ
AP
3-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
UG - 200 cm/s
9.5 i/n'
7.5 cm W.C.
dfe = 3.8 cm
HS 15 cm
RED IRON OXIDE AEROSOL
_L
_L
0.2 O.S I 5
AERODYNAMIC PARTICLE DIAMETER, umA
Figure C-27. Experimental grade pcnctrntion
curves.
ID
224
-------
1.0
0.1
0.05
0.01
RUN NO. C3-24
C3-2S
3-STACE MOBILE BED
HARDWARE SCREEN SUPPORT
u.. 200 cm/s
QL/QG 12.9 i/m1
AP,, = 9.0 cm W.C.
H
3.8 cm
IS cm
RED IRON OXIDE AEROSOL
I
0.2 0.5 1 S
AERODYNAMIC PARTICLE DIAMETER, um
10
Figure C-28. Experimental grade penetration
curves .
1.0
0.5
0.05
RUN NO. C3-26
C3-27
3-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
ur 230 cm/s
b
QL/QG = 4.9 l/m!
4Pw = 7.5 cm W.C.
d. 3.8 cm
D
HS 15 cm
RED IRON OXIDE AEROSOL
I
0.2 0.5 1 5 10
AERODYNAMIC PARTICLE DIAMETER, umA
Figure C-29. Experimental grade penetration
curves.
0.5
0.05
0.01
RUN NO. C3-28
C3-29
3-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
uf = 230 cm/s
QL/qG = 8.2 (./Hi3
iP =8.9 cm W.C.
d. = 3.8 cm
H5 « 15 cm
RED IRON OXIDE AEROSOL
, . I . , . , I f
I
0.2 O.S 1 5
AERODYNAMIC PARTICLE DIAMETER, unA
Figure C-30. Experimental graile penetration
curves.
0.5
0.01
RUN NO. C3-30
C3-31
3-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
UG 230 cm/s
QL/QG =11.2 i/m1
APu - 10 cm W.C.
d, • 3.8 cm
b
HS = 15 cm
RED IRON OXIDE AEROSOL
0.2 0.5 1 5
AERODYNAMIC PARTIC1.I; DIAMETER, umA
Figure C-31. Ir,xper imental sjratle pcneT r;i t i on
curves.
225
-------
1.0
0.5
0.05
0.01
RUN NO. C3-32
C3-33
3-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
UG = 230 cm/s
QL/QG =4.9 H/rn'
APW = 9.8 cm W.C.
dfc = 3.8 cm
Hs 23 cm
RED IRON OXIDE AEROSOL
. , I
0.2 0.5 1 S
AERODYNAMIC PARTICLE DIAMETER, ymA
Figure C-32. Experimental yrade penetration
curves.
1.0
0.5
0.1
0.05
0.01
RUN NO. C3-34
C3-35
3-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
u,
G
230 cm/s
8.2 J/m3
APw 10.9 cm W.C.
d^ 3.8 cm
HS 23 cm
RED IRON OXIDE AEROSOL
! . I . ...I
0.2 0.51 510
AERODYNAMIC PARTICLE DIAMETER, umA
Figure C-33. Experimental grade penetration
curves.
0.5
0.1
0.01
i ' I ' " M
RUN NO. C3-36
C3-37
3-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
Ug 230 cm/s
QL/QG =11.2 l/m3
iPw 13.1 cm W.C.
db 3.8 cm
Hs 23 cm
RED IRON OXIDE AEROSOL
I .
0.2 0.5 1 5 10
AERODYNAMIC PARTICLE DIAMETER, umA
Figure C-34. Experimental grade penetration
curves.
0.5
0.05
_ RUN NO. C3-38
3-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
UG 260 cm/s
QL/QG 4.4 JiVm3
4PW 11.1 cm W.C.
d^ = 3.8 cm
HS • 23 cm
RED IRON OXIDE AEROSOL
_L
0.2 0.5 1 5 10
AERODYNAMIC PARTICLE DIAMETER, umA
Figure C-35. Experimental s^rade penetration
curve.
226
-------
0.1
0.05
RUN NO. C3-39
C3-40
3-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
Up 230 cm/s
n,/Qp 11.2 i/m3
AP^ 13.1 cm W.C.
dfc 3.8 cm
>L - 23 cm
FLY ASH AtROSOL
0.2 0.5 1 5
AERODYNAMIC PARTICLE DIAMETER, umA
10
0.1
M 0.05
RUN NO. C3-41
C3-42
S-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
ufi - 230 tm/5
QL/QR 4.9 Jl/m3
AP -9.6 cm W.C.
HS - 23 cm
FLY ASH AEROSOL
0.2 O.S 1 5
AERODYNAMIC PARTICLE DIAMETER, jimA
Figure C-36. Experimental grade penetration
curves.
Figure C-37. Experimental grade penetration
curves.
0.5
0,1
*• 0.05
a
0.01
\ I I I Mil
RUN SO. C3-43
C3-44
3-STA6E MOSILl BID
" HARDWARE SCR1EN SUPPORT
u, - 200 cm/s
3
APW U.8 Cm W.C.
db - 3.8 c«i
M, « 23 em
FLY ASH AEROSOL
O.S
1 5 10
PARTICLE DIAMETER, umA
C-SS. Experifnental grade penstration curves.
1.0
0.5
0.1
0.05
0.01
RUN NO. C3-45
C3-46
3-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
Ur ' 290 cm/s
QL/QO * 6-5 l/m3
APW = 11.6 OS W.C.
4b " 3.8 cm
HS • 23 em
FLY ASH AEROSOL
-•I i i i
0.2 O.S 1 5
AERODYNAMIC PARTICLE DIAMETER, umA
Figure 39, Experii1i§ntal grade penetration
eurves.
_LL
10
227
-------
1.0
0.5
o
H
0.1
0.05
0.01
RUN NO. C3-47
C3-
3-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
UG « 290 cm/s
QL'QG " 3-9 t^a*
APW - 10.2 cm W.C.
d^ • 3.8 cm
H5 • 23 cm
CUPOLA DUST AEROSOL
I I I I
I
I i I I I I
0.2 0.5 1 5
AERODYNAMIC PARTICLE DIAMETER, [imA
10
1.0
0.5
0.05
0.01
RUN 110. C3-49
3-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
UG 290 cm/s
QL/QG = 6.5 Z/m'
AP - 11.6 cm W.C.
dfc - 3.8 cm
Hs • 23 cm
CUPOLA DUST AEROSOL
I
I I I I I I I
0.2 0.5 1 5
AERODYNAMIC PARTICLE DIAMETER, umA
10
Figure C-40. Experimental grade penetration curves.
Figure C-41. Experimental grade penetration curve.
I
I
1.0
0.5
0.1
0.05
0.01
RUN NO. C3-50
C3-51
. 3-STAGE MOBILE BED
. HARDWARE SCREEN SUPPORT
UG - 290 cm/s
- QL/QG - 8.8 */m>
APW - 14 cm W.C.
d. - 3.8 cm
HS - 23 cm
CUPOLA DUST AEROSOL
0.2 0.5 1 5
AERODYNAMIC PARTICLE DIAMETER, umA
0.5
0.1
0.05
0.01
\ I I I I I I
t III I I I 1
RUN NO. C3-52
C3-53
3-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
UG = 230 cm/s
QL/QG -11.2 l/m'
APW - 13.6 cm W.C.
d. > 3.8 cm
CUPOLA DUST AEROSOL
I
I I I l l I
I i I
0.2 0.5 1. 5
AERODYNAMIC PARTICLE DIAMETER, ymA
10
Figure C-42. Experimental grade penetration curves.
Figure C-43. Experimental grade penetration curves.
228
-------
1.0
0.5
0.05
"1 1 I I I I
RUN NO. C3-54
C3-55
C3-56
C3-S7
i—i—r i MIL
3-STAGE MOBILE BED
- HARDWARE SCREEN SUPPORT
UG • 390 cm/s
- QL/QG - 4.8 l/m'
iPw - 12.5 cm W.C.
d, = 3.8 cm
CUPOLA DUST AEROSOL
I I I I I I I I
I
I I 1 I I I I
0.2 0.5 1 5 10
AERODYNAMIC PARTICLE DIAMETER, pmA
1.0
0.5 -
0.1 -
0.05 -
0.01
3-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
- QL/QG = 6.6 I/in3
AP = 14.3 cm W.C.
0.2 0.5 1 5
AERODYNAMIC PARTICLE DIAMETER, umA
Figure C-44. Experimental grade penetration curves.
Figure C-45. Experimental grade penetration curves.
1.0
0.5
0.05
0.01
RUN NO. C3-60
_ 3-STAGE MOBILE BED
- HARDWARE SCREEN SUPPORT
- QL/Q
- 390 cm/s
2.9 */m°
•I've
4PW - 9.7 cm W.C.
d. 3. 8 cm
HS 28 cm
CUPOLA DUST AEROSOL
I I I I I I I I
I i i I 1 I i i
0.2 0.5 1 5 10
AERODYNAMIC PARTICLE DIAMETER, umA
1.0
0.5
—i—r i i i i 11
- RUN NO. C3-61
C3-62
0.1
0.05
0.01
~i 1—r i i i i
3-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
UG = 390 cm/s
QL/QG =7.5 */m!
APW = 15.4 cm W.C.
db = 3.8 cm
Hs = 23 cm
CUPOLA DUST AEROSOL
J 1
0.2 0.5 1 5 10
AERODYNAMIC PARTICLE DIAMETER, pmA
Figure C-46- Experimental grade penetration curve.
Figure c-47. Experimental grade penetration curves.
229
-------
1.0
0.5 h
0.01
RUN NO.C3-63
C3-64
3-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
ur = 390 cm/s
QL/QG *-s n/1"3
4P-, - 17.S cm W.C.
dj '= 3.8 cm
i H, = 23 cm
H CUPOLA DUST AEROSOL
0.2 0.5 1 S 10
AERODYNAMIC PARTICLE DIAMETER, umA
0.5
0.1
0.05
0.01
RUN NO. C3-65
C3-66
h 3-STAGE MOBILE BED
|- HARDWARE SCREEN SUPPORT
440 cm/s
2.5 £/m3
AP« 10.3 cm W.C.
3.8 cm
I ns - 23 cm
f" CUPOLA DUST AEROSOL
I i I I I I I I
0.2 0.5 1 5 10
AERODYNAMIC PARTICLE DIAMETER, umA
Figure C-48. Experimental grade penetration curves.
Figure C-49. Experimental grade penetration curves.
1.0
0.5 \-
0.1
0.05
0.01
RUN NO. C3-67
C3-68
3-STAGE MOBILE BED
t- HARDWARE SCREEN SUPPORT
up, 410 cm/s
|- ""AP'w = S.6 cm W.C.
db 3.8 cm
I I I I I I I
0.2 0.5 1 5 10
AERODYNAMIC PARTICLE DIAMETER, umA
0.1
0.05
- 3-STAGE MOBILE BED
i- HARDWARE SCREEN SUPPORT
uQ 410 cm/s
QL/QG 4.5 l/m1
4PW 12.2 cm W.C.
dh - 3.8 cm
Hs = 23 cm
0.2 0.5 1 5 10
AERODYNAMIC PARTICLE DIAMETER, umA
Figure C-50. Experimental grade penetration curves.
I-igure C-51, Experimental grade penetration curves.
230
-------
0.5
S 0.1
1
w
a
w 0.05
0.01
n i i iij
RUN NO. C3-71
C3-72
C3-73
3-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
ur 410 cm/s
QL/qg - 6-2 1/ml
4PW = 14.3 cm W.C.
dh 3.8 cm
H= 23 cm
l I I I I I I I
0.2 0.5 1 5 10
AERODYNAMIC PARTICLE DIAMETER, pmA
1.0
0.5
0.1
0.05
0.01
RUN NO. C3-74
3-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
UG • 230 cm/s
QL/QG 6.5 1/m3
4P = 9.7 cm W.C.
Hs 23 cm
CUPOLA DUST AEROSOL
1 1 I I I
I
I I I 1 I I I
0.2 0.5 1 5
AERODYNAMIC PARTICLE DIAMETER, ymA
10
Figure C-52. Experimental grade penetration curves.
Figure C-53. Experimental grade penetration curve.
1.0
0.5
0.1
0.05
0.01
RUN NO. C3-75
3-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
UG 290 cm/s
Q- = 7.7 l/m3
10.1 cm W.C.
4P
HS 23 cm
CUPOLA DUST AEROSOL
l l l l l l I
0.2 0.5 1 5 10
AERODYNAMIC PARTICLE DIAMETER, umA
0.5 -
0.1 .
3-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
0.2 0.51 510
AERODYNAMIC PARTICLE DIAMETER, umA
Figure C-54. Experimental grade penetration curve.
Figure C-55. Experimental grade penetration
curves.
231
-------
0.01
RUN NO. C3-78
C3-79
3-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
UG 410 cm/s
QL/QG - 5.4 l/m'
AP^ 10.7 cm W.C.
d. 5.1 cm
Hs 23 cm
CUPOLA DUST AEROSOL
I I I I i I I I 1__
0.2 0.5 1 5 10
AERODYNAMIC PARTICLE DIAMETER, umA
RUN NO. C3-80
C3-81
|- 3-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
Uj, 410 cm/s
QL/QG 7.0 i/m'
ipj 13.2 cm W.C.
dfc 3.8 cm
HS 23 cm
CUPOLA DUST AEROSOL
I I I I I I
0.5 1 5 10
AERODYNAMIC PARTICLE DIAMETER, pmA
Figure C-56. Experimental grade penetration
curves.
F igur e C- 5 7
Experimental grade penetration
curves.
1.0
0.1
0.01
' -
RUN NO. C3-82
3-STAGE MOBILE BED
PLASTIC NET SUPPORT
UG 420 cm/s
QL/QG 7.2 I/],'
AP = 39.5 cm W.C.
d, 3.8 cm
Hs 23 cm
CUPOLA DUST AEROSOL
I L I 1 I i I I 1
0.5 1
AERODYNAMIC PARTICLE DIAMETER,
0.05
3-STAGE MOBILE BED
PLASTIC NET SUPPORT
UG 420 cm/s
qL/QQ 6.4 i/m3
APW 33 cm W.C.
db 3.8 cm
H5 23 cm
CUPOLA DUST AEROSOL
0.1 0.5 1
AERODYNAMIC PARTICLE DIAMETER,
Figure C-58. Experimental grade penetration
curve.
Figure C-59. Experimental grade penetration curves.
232
-------
0.5
0.01
RUN NO. C3-87 -
C3-88
3-STAGE MOBILE BED
PLASTIC NET SUPPORT
UG 420 cm/s
Q]/QG 5.6 i/m3
APU 22.4 cm W.C.
d, 3. 8 cm
HS 23 cm
CUPOLA DUST AEROSOL
0.5 1
AERODYNAMIC PARTICLE DIAMETER,
0.1
« 0.05
- RUN NO. C3-S9
C3-90
I r [ i [
3-STAGE MOBILE BED
PLASTIC NET SUPPORT
"G 410 cm/s
QL/QG 5.6 H/m!
AP,, 25.7 cm W.C
3.8 cm
HS = 23 cm
CUPOLA DUST AEROSOL
I
I
0.1 0.51
AERODYNAMIC PARTICLE DIAMETER, ymA
Figure C-60. Experimental grade penetration curves.
Figure (. 61. Experimental grade penetration curves.
0.5
0.1
0.05
0.01
RUN NO. C3-91
C3-92
3-STAGE MOBILE BED
PLASTIC NET SUPPORT
VIG
4Pw
db
H =
410 cm/s
4.8 1/m3
18.3 cm
3.8 cm
23 cm
W.C.
CUPOLA DUST AEROSOL
I I I
I
I
0.1 0.5 1 5
AERODYNAMIC PARTICLE DIAMETER, pmA
RUN NO. C3-93
C3-94
_ 3-STAGE MOBILE BED
_ PLASTIC NET SUPPORT
210 cm/s
9.3 i/m'
13.8 cm W.C.
3.8 cm
- QL/Q
LG
AP,
CUPOLA DUST AEROSOL
0.01 i I I I I I I
I
I I I I TTL
0.3 1 5 10
AERODYNAMIC PARTICLE DIAMETER, umA
Figure C-02. Experimental grade penetration curves.
Figure C-d3. Experimental grade penetration
curves.
233
-------
RUN \0. C3-95
C3-96
_ 3-STAGE MOBILE BED
_ PLASTIC NET SUPPORT
ur 210 cm/s
4PW 15.3 cm W.C.
d. = 3.8 cm
b
HS • 23 cm
CUPOLA DUST AEROSOL
I 1 I
0.1 O.S 1
AERODYNAMIC PARTICLE DIAMETER,
0.5
c
o
u
n
t-
M-i
O
i "-1
UJ
z
UJ
* 0.05
u
o£
&.
0.01
1 1 1 1 1 1 1 1 I ,. i
RUN MO. C3-97 \
; \
\
-
- 3-STAGE MOBILE BED
- PLASTIC NET SUPPORT
UG 210 cm/s
" QL/QG 12.6 l/m1
APw 16.8 cm K.C.
d^ 3.8cm
HS 23 cm
CUPOLA DUST AEROSOL
I 1 1 1 1 1 1 1 1 1
:
.
:
-
-
0.1 0.5 1
AERODYNAMIC PARTICLE DIAMETER, ymA
Figure C-64. Experimental grade penetration curves.
Figure C-65. Experimental grade penetration curve.
0.1
0.01
RUN NO. C3-98
3-STAGE MOBILE BED
PLASTIC NET SUPPORT
UG •= 300 cm/s
QL/QG ** 10.7 l/m1
APw = 25.4 cm W.C.
db 3.8 cm
HS 23 cm
CUPOLA DUST AEROSOL
0.1 0.5 1
AERODYNAMIC PARTICLE DIAMETER, pmA
0.5
0.05
RUN NO. C3-99
J-STAGE MOBILE BED
PLASTIC NET SUPPORT
UG ISO cm/s
QL/Q0 6.3 l/m3
APK = 18.3 cm W.C.
d. = 3.8 cm
HS 23 cm
CUPOLA DUST AEROSOL
0.2 0.5 1 5
AERODYNAMIC PARTICLE DIAMETER, u
10
figure C-66. Experimental grade penetration curve.
Figure C-67. Experimental grade penetration curve.
-------
APPENDIX "D"
SLURRY SCRUBBING PENETRATION CURVES
235
-------
1.0
o.s
0.1
0.05
0.01
3-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
UG = 280 cm/s
QL/QG = 6.7 l/m3
APw 12 cm W.C.
db 3.8 cm
HS = 23 cm
CUPOLA DUST AEROSOL
0.2 0.51 5
AERODYNAMIC PARTICLE DIAMETER, ymA
Figure D-l. Experimental penetration curves.
10
1.0
0.5
0.1
0.05
0.01
RUN NO. S3-2
S3-7
3-STAGE MOBILE BED
HARDWARE SCRE-EN SUPPORT
UG 280 cm/s
QL/QG 9.1 Jl/m3
AP 14.0 cm W.C.
db 3.8 cm
H 23 cm
CUPOLA DUST AEROSOL
I L I I ... I ,
I
0.2 0.5 1 5 10
AERODYNAMIC PARTICLE DIAMETER, umA
Figure D-2. Experimental penetration curves.
1.0
0.5
0.1
0.05
0.01
RUN NO. S3-3
S3-
3-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
UG • 340 cm/s
QL/QG = 5.5 H/m3
6PW = 12.3 cm W.C.
db = 3.8 cm
HS 23 cm
CUPOLA DUST AEROSOL
I
0.2 0.5 1 5
AERODYNAMIC PARTICLE DIAMETER, umA
Figure D-3. Experimental penetration curves.
1.0
0.5
0.1
m 0.05
0.01
RUN NO. S3-4
S3-9
3-STAGE MOBILE BED
HARDWARE SCREEN SUPPORT
UG 340 cm/s
QL/Q_ 7.6 H/m3
APW 14.2 cm W.C.
d^ - 3.8 cm
HS 23 cm
CUPOLA DUST AEROSOL
—I 1 1 I I I I I L_
_L
0.2 0.5 1 5 10
AERODYNAMIC PARTICLE DIAMETER, vimA
Figure D-4. Experimental penetration curves.
236
-------
APPENDIX "E"
GRADE PENETRATION CURVES FOR F/C RUNS
237
-------
100
90
80
70
60
50
30
10
RUN NO. FC-1
FC-2
FC-3
THREE STAGE MOBILE BED
HS 23 cm
d, 3.8 cm
D
QL = 273 l/rain
q1 = 0
AP 12.6 cm W.C.
CUPOLA DUST AEROSOL
0.5
,
Figure E-l. The penetration curve for
RunsNo. FC-1, FC-2 and
FC-3.
30
20
10
i ii—n~
FC-5
THREE STAGE MOBILE BED
23 cm
3.8 cm
273 i/min
0.072 j>/g D.n.
12.6 cm W.C.
CUPOLA DUST AEROSOL
0.6 0.8 1
Figure'F,-2. The penetration curve for
Runs No. l;C-4 and FC-S.
100
90
80
70
60
50
40
10
RUN NO. FC-8
q' = 0.12
THREE STAGE
MOBILE BED
23 cm W.C.
3.8 cm W.C.
AP 12.6 cm W.C.
CUPOLA DUST AEROSOL
0.6
RUN NO. FC-7
0.11
V
urn A
100
90
80
70
60
50
40
30
20
10
RUN NO. PC-6
CUPOLA DUST AEROSOL
0.4
0.6 0.8 1
V
Figure E-4. The penetration curve for Runs
No. FC-7 and FC-8.
Figure E-3. The penetration curve for Run
No. FC-6.
238
-------
100
90
80
70
60
50
40
30
20
10
0.4
RUN NO. FC-9
FC-10
CUPOLA DUST AEROSOL
0.6 0.8 1
,
Figure E-S.
The penetration curve for Runs
No. FC-9 and FC-10.
100
90
80
70
60
50
40
30
20
10
RUN NO. FC-11
q' 0.16
THREE STAGE
- MOBILE BED
AP 12.9 cm W.C.
CUPOLA DUST AEROSOL
0.4
0.6 0.8 1
umA
Figure I:-6. The penetration curve for Runs
No. FC-11, FC-12 and FC-13.
100
90
80
70
60
50
40
30
20
10
RUN NO. FC-14
q' = 0.18
0.18
THREE STAGE
MOBILE BED
H 23 cm
d. 3 . 8 cm
D
Q 273 l/min
&P 12.6 cm W.C.
CUPOLA DUST AEROSOL
0.4
0.6 0.8 1
d , ymA
Figure
The penetration curve for Runs
No. FC-1-1. PC-IS and FC-16.
100
90
80
™
60
50
40
30
20
10
I 1 T
-r-r
THREE STAGE
MOBILE BED
UN NO. FC-17
q' = 0.18
HS 23 cm
dfe = 3.8 cm
QL = 273 d/min
AP = 12.6 cm W.C.
CUPOLA DUST AEROSOL
0.4
0.6
,
Figure Ii-8. The penetration curve for
Runs No. FC-17 and l-C-20.
239
-------
ion
90
80
70
60
50
30
RUN NO. FC-19
q1 0.15
IN NO. FC-18
= o.ir
AP = 13.5 en W.C.
CUPOLA DUST AEROSOL
0.4
0.6 0.8 I
,
ymA
100
90
80
70
60
SO
20
10
-1 1—I—1—1-
THREE STAGE
MOBILE BED
RUN NO. FC-21
q' = 0.16
.UN1 NO. FC-ZZ
0.13-
HS = 23 cm
d, • 3.8 cm
D
QL = 273 2/min
AP = 13.5 cm W.C.
CUPOLA DUST AEROSOL
0.4
0.6 0.8 1
Figure Ii-9. The penetration curve for
Runs No. FC-18 and FC-19.
Figure l!-10. The penetration curve for
Runs No. FC-21 and FC-22.
100
70
20
10
RUN NO. FC-Z4
q1 0.23
THREE STAGE
MOBILE BED
23 cm
3. 8 cm
Q. Z73 £/min
AP 14.4 cm W.C.
CUPOLA DUST AEROSOL
RUN NO. FC-23
q' = 0.24
RUN NO.
q'
FC-25-
0.25
100
0.6 0.8 1
d , umA
0.4
0.6 0.
V
Figure E-ll. The penetration curves for
Runs No. FC-23, FC-24 and
FC-25.
Figure E-12. The penetration curves for
Runs No. FC-2^ and FC-27.
240
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing;
in n r r
4. TITLE AND SUBTITLE
Mobile Bed Flux Force/Condensation Scrubbers
3. RECIPIENT'S ACCESSION NO
5. REPORT DATE
February 1979
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO.
S.C.Yung, R.Chmielewski, andS.Calvert
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Air Pollution Technology, Inc.
4901 Morena Boulevard, Suite 402
San Diego, California 92117
10. PROGRAM ELEMENT NO.
EHE624A
11. CONTRACT/GRANT NO.
68-02-2124
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; 11/75 - 12/78
14. SPONSORING AGENCY CODE
EPA/600/13
is.SUPPLEMENTARY NOTES J.ERL-RTP project officer is Dale L. Harmon, MD-61, 919/541-
2925.
16. ABSTRACT
The report gives results of an experimental determination of fine particle
collection in mobile bed scrubbers. Particle collection efficiency increased greatly
as the gas-phase pressure drop increased. With no water vapor condensation, the
performance capability of a mobile bed scrubber is less than that of a gas-atomized
spray scrubber with the same pressure drop. Compared to packed bed and sieve
plate scrubbers, the mobile bed scrubber has better efficiency when the pressure
drop is above 20 cm W. C. Limestone in the scrubber liquid has no effect on particle
collection. When the mobile bed scrubber was used as a flux force/condensation
(FF/C) scrubber, it had better performance characteristics than sieve plate and
spray scrubbers with condensation. However, its capability is inferior to a FF/C
system consisting of a condenser and venturi scrubber. Design equations reported in
the literature are inadequate to predict the collection efficiency and pressure drop
of the mobile bed scrubber. The study developed new correlations to predict particle
collection and pressure drop.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COS AT I Field/Group
Pollution
Scrubbers
Dust
Aerosols
Flux Density
ondensing
Pollution Control
Stationary Sources
Mobile Bed Scrubbing
Particulate
Flux Force/Condensa-
tion
13 B
07A,13I
11G
07D
14B
3. DISTRIBUTION STATEMENT
19. SECURITY CLASS (ThisReport)
Unclassified
21. NO. OF PAGES
261
Unlimited
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
241
------- |