EPA-600/2-76-200
July 1976
Environmental Protection Technology Series
                   STUDY OF  HORIZONTAL-SPRAY
        FLUX  FORCE/CONDENSATION  SCRUBBER
                                Industrial Environmental Research Laboratory
                                     Office of Research and Development
                                    U.S. Environmental Protection Agency
                               Research Triangle Park, North Carolina 27711

-------
               RESEARCH REPORTING SERIES

Research reports of the Office of Research and Development, U.S. Environmental
Protection  Agency, have been grouped into five series. These five broad
categories were established to facilitate further development and application of
environmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The five series are:

     1.    Environmental Health Effects Research
     2.    Environmental Protection Technology
     3.    Ecological Research
     4,    Environmental Monitoring
     5.    Socioeconomic  Environmental Studies

This report has been assigned  to the ENVIRONMENTAL PROTECTION
TECHNOLOGY series. This series describes research performed to develop and
demonstrate instrumentation, equipment, and methodology to repair or prevent
environmental degradation  from point and non-point sources of pollution. This
work provides  the new or  improved technology required for the control and
treatment of pollution sources to meet environmental quality standards.
                    EPA RE VIEW NOTICE

This report has been reviewed by  the U.S.  Environmental
Protection Agency, and approved for publication.  Approval
does not signify that the contents necessarily reflect the
views and policy of the Agency, nor does mention of trade
names or commercial products constitute endorsement or
recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.

-------
                                      EPA-600/2-76-200
                                      July 1976
                  STUDY OF

            HORIZONTAL-SPRAY

FLUX  FORCE/CONDENSATION SCRUBBER
                        by

        Seymour Calvert and Shui-Chow Yung

                  A.P.T. , Inc.
          4901 Morena Boulevard (Suite 402)
               San Diego, CA 92117
          Contract No.  68-02-1328, Task 10
               ROAPNo.  21ADL-002
            Program Element No.  1AB012
          EPA Task Officer:  L.E. Sparks

    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

-------
                           FOREWORD

     This report, "Study of Horizontal Spray Flux Force/Conden-
sation Scrubber," is the final report submitted to the Industrial
Environmental Research Laboratory for E.P.A. Contract No. 68-02-
1328, Task No. 10.
     The principal objective of this program was to experimen-
tally evaluate fine particle collection in a laboratory pilot
scale flux force/condensation (FF/C) scrubber and to determine
the feasibility of application of FF/C scrubbing to industrial
sources.  The main activities under the scope or work were:
     Based on the results of the previous theoretical and experi-
mental study of FF/C scrubbing;
     1.  Design and fabricate the pilot scale horizontal spray
        scrubbers large enough for the exploration of scale-up
        problems.
     2.  Conduct a laboratory pilot experimental program to:
        A. Determine feasibility for fine particle collection.
        B. Develop design equations and scale-up criteria.
        C. Determine optimum operating conditions for FF/C
           scrubbing.
        D. Investigate potential operational and maintenance
           problems.
        E. Determine effects of particle size distribution
           on the performance of FF/C scrubbers.
     3.  Prepare revised engineering and cost analysis to in-
        corporate results of the experimental study.
     4.  Recommend a detailed industrial pilot test program.
     Dr. Leslie E. Sparks of the Industrial Environmental Re-
search Laboratory, U.S. Environmental Protection Agency was
the Project Officer for this program.
                               111

-------
     Dr.  Seymour Calvert of Air Pollution Technology,  Inc.  was
the Project Director.
                                IV

-------
                        ABSTRACT

      This report presents the results of a laboratory
pilot scale evaluation of a Flux Force/Condensation
(FF/C) scrubber  for the collection of fine particles.
FF/C scrubbing includes the effects of diffusiophoresis,
thermoporesis, Stefan  flow and particle  growth due to
the condensation of water vapor.   Fine particles  are  those
smaller than 2 microns in diameter.
      The  FF/C scrubber tested was of horizontal spray
configurations.  Effects of the scrubber configurations,
liquid and gas flowrates, particle number concentration
and  the amount of vapor condensation were studied experi-
mentally.  Fractional particle penetrations were measured
with  cascade  impactors and are presented in terms of
particle  penetration as a function of particle size.   The
experimental  results are compared with predictions from
mathematical models.
     Optimum operational regions and the technical and
economic feasibility of FF/C scrubbing are determined and
demonstrated for a fine particle pollution source.  It was
confirmed that FF/C scrubbing is an attractive control
method for fine particles when high efficiency is required
or when the gas is hot enough to evaporate the necessary
water vapor for condensation in the scrubber.   A program
to demonstrate FF/C scrubbing at pilot scale for the  con-
trol of fine particulate emissions from industrial sources
is described.
      This report was submitted by Air Pollution Technology,
Inc.  in fulfillment of Contract No. 68-02-1328, Task  No.  10,
under the sponsorship of the Environmental Protection Agency,
Work  was  completed  on December 14, 1974.
                           v

-------
                       CONTENTS


                                                  Page

Foreword 	 	 ..... iii

Abstract ...... 	 	  v

List of Figures	

List of Tables	

Nomenclature	

Acknowledgements 	 xiii

     1. Introduction	1

     2. Summary, Conclusions and Recommendations  . 3

     3. Background	12

     4. Experimental Pilot Plant 	 15

     5. Experimental Results and Discussions ... 30

     6. FF/C Scrubber Performance Prediction
        Methods	41

     7. Economic Feasibility	51

     8. Future Research Recommendations  	 66

References	82

Appendix 1	84
                          VI

-------
                     LIST OF FIGURES
Number                    •                       Page

 4-1   Process flow sheet for the FF/C spray
       scrubbing system. .  .  .i. .  .  . .  ......   16

 5-1   Penetration versus condensation ratio,
       one stage spray  . .' .  .   .':•;;  . ....>.  .32

 5-2   Penetration versus condensation ratio,
       one stage spray  . i  ...   . .  .	'33

 5-3   Penetration versus condensation ratio,
       one stage spray	   34

 5-4   Penetration versus condensation ratio,
       one stage spray	   35

 5-5   Penetration-versus condensation ratio,   .. •.;
       three stage co-current spray  ......   36

 5-6   Penetration versus condensation ratio,
       three stage counter-current spray ....   37

 5-7   Comparison of FF/C plate and spray
       scrubber results for 1 ymA particles  .   .   38

 6-1   Predicted,particle collection.efficiency
       for sprays by inertial impaction and in-
       terception. . . .   . .  .	••.... .   .   42

 6-2   Scrubber area covered  by sprays	   42

 6-3   Spray scrubber penetration predictions
       for 500 ym drop diameters	   45

 6-4   Spray scrubber penetration predictions
       for 300 ym drop diameters	   45

 6-5   Experimental  1.0 ymA particle penetration
       for spray scrubber	   47

 6-6   Experimental  1.0 ymA particle penetration
       for spray scrubber	   47
                          vn

-------
                       LIST OF FIGURES


Number                                                     Page

 6-7   Experimental 1.0 ymA particle penetration for spray
       scrubber	47

 7-1   Operating cost comparison of FF/.C and H.E. scrub-
       bers 	53
                                             1 ,.
 7-2   Process diagram for cupola gas cleaning  ...... 61

 8-1   Typical process design of a FF/C scrubber system .   . 72
Appendix . .
1
1
1
1
1
1
1
1
1
1
1
-1
-2
-3
-4
-5
-6
-7
-8
-9
-10
-11
Particle penetration versus
single stage spray scrubber
Particle penetration versus
single stage spray scrubber
Particle penetration versus
single stage spray scrubber
Particle penetration versus
single stage spray scrubber
Particle penetration versus
single stage spray scrubber
Particle penetration versus
single stage spray scrubber
Particle penetration versus
single stage spray scrubber
Particle penetration versus
single stage spray scrubber
Particle penetration versus
single stage spray scrubber
Particle penetration versus
single stage spray scrubber
Particle penetration versus
sinele staee snrav scrubber
aerodynamic
, runs 1 and
aerodynamic
, runs 3 and
aerodynamic
, runs 8 and
aerodynamic
, runs 13, 14
aerodynamic
, runs 17 and
aerodynamic
, runs 21, 22
aerodynamic
, runs 24, 25
aerodynamic
, runs 29 and
aerodynamic
, runs 32 and
aerodynamic
, run 35
aerodynamic
. run 38 . .
diameter,
2 ....
diameter,
4 . ....
diameter,
10 .. . .
diameter ,
and 16
diameter,
19. .
diameter,
and 23
diameter,
and 28 .
diameter ,
31 ...
diameter ,
34 ...
diameter ,
diameter ,
94
94
94
. 94
95
. 95
. 95
. 95
. 96
. 96
, 96
                            Vlll

-------
                        LIST OF FIGURES .

Number                                                    Page

  Appendix

  1-12  Particle penetration versus aerodynamic diameter,
        single stage spray scrubber, .runs 37 and 39 ...;.  . .  96

  1-13  Particle penetration versus aerodynamic diameter,
        one stage, run 42	  97

  1-14  Particle penetration versus aerodynamic diameter,
        one stage, runs 43 and 45 . .  .  .  .  .  . .  '. .   .  . .97

  1-1.5  Particle penetration versus aerodynamic diameter,
        one stage, runs 46 and 50 ..	97

  1-16  Particle penetration versus aerodynamic diameter,
        single stage,  run 48	  97

  1-1-7  Particle penetration versus aerodynamic diameter,
        one stage, runs 51 and 54	98

  1-18  Particle penetration versus.aerodynamic diameter,
        one stage, runs 55 and 56 . .  .  .  .  .  .-.  . .   .  . .  98

  1-19  Particle penetration versus aerodynamic diameter,
        one stage, 'runs 59 and 60	  98

  1-20  Particle penetration versus aerodynamic .diameter,
        one stage, runs 61 and 63	 .  98

  1-21  Particle penetration versus aerodynamic diameter,
        one stage, runs 64 and 66	  99

  1-22  Particle penetration versus aerodynamic diameter,
        one stage, runs 67 and 71	  99

  1-23  Particle penetration versus aerodynamic diameter,
        one stage, runs 70 and 72 .'...'	-99

  1-24  Particle penetration versus aerodynamic diameter,
        3 stage co-current spray scrubber, runs 75, 77
        and 79	99

  1-25  Particle penetration versus aerodynamic diameter,
        3 stage co-current spray scrubber, runs 80 and 81 .  100
                                 IX

-------
                        LIST OF FIGURES


Number                                                    Page

  Appendix

  1-26  Particle penetration versus aerodynamic diameter,
        3 stage co-current spray scrubber,:runs 85,  86
        and 88	100

  1-27  Particle penetration versus aerodynamic diameter,
        3 stage co-current spray scrubber,  runs 89,  91
        and 92	100

  1-28  Particle penetration versus aerodynamic diameter,
        3 stage co-current spray scrubber,  runs 95 and 96.  100

  1-29  Particle penetration versus aerodynamic diameter,
        3 stage counter-current, runs 99 and 101	101

  1-30  Particle penetration versus aerodynamic diameter,
        3 stage counter-current, runs 102 and 103  ....  101

  1-31  Particle penetration versus aerodynamic diameter,
        3 stage counter-current, runs 104 and 105  ....  101

  1-32  Particle penetration versus aerodynamic diameter,
        3 stage counter-current, runs 108 and 109  ....  101

  1-33  Particle penetration versus aerodynamic diameter,
        3 stage counter-current, runs 112 and 113  ....  102

  1-34  Particle penetration versus aerodynamic diameter,
        3 stage counter-current, runs 121,  122 and 123 .  .  102

  1-35  Particle penetration versus aerodynamic diameter,
        3 stage counter-current, runs 114,  115 and 117 .  .  102

-------
                      LIST OF TABLES

Number

 2-1   Major Industrial Particulate Sources for
       Which FF/C Scrubbing is Attractive ....   4

 4-1   Equipment Specifications .........  17

 4-2   Stream Flow Rates of the Spray Scrubbing
       System 	  ........  18

 4-3   Spray Scrubber Operation Modes	  24

 6-1   FF/C Spray Scrubber Design Equations ...  49

 7-1   Gas Conditions and Fan Power Costs ....  55

 7-2   Typical Penetrations for Plates and Spray.  55

 7-3   Cost Comparison of Cupola Emission Control
       Systems	..'...-  64

 8-1   Estimated Schedule of Performance  ....  76

 8-2   Detailed Cost Breakdown  .  .  . .  .  . .  .  .  79

 Appendix                               .   .
 1-1   Horizontal Spray Scrubber;  Operating Con-
       ditions and Performance (Runs  1-23).  ...  85

 1-2   Horizontal Spray Scrubber;  Operating Con-
       ditions and Performance (Runs  2.4-40)  ...  87

 1-3   Horizontal Spray Scrubber;  Operating Con-
       ditions and Performance (Runs  41-50)  ...  88

 1-4   Horizontal Spray Scrubber;  Operating Con-
       ditions and Performance (Runs  51-74)  ...  88

 1-5   Horizontal Spray Scrubber;  Operating Con-
       ditions and Performance (Runs  75-96)  .  .' .  90

 1-6   Horizontal Spray Scrubber;  Operating Con-
       ditions and Performance (Runs  97-123)   .  .  91
                           XI

-------
                       NOMENCLATURE

A    = constant
B    = constant
c    = particle mass concentration, g/DNm3 gas
C'   = Cunningham correction factor, dimensionless
d    = diameter, cm, ym or ymAE ym  (g/cm3) 1/2
d    = mass mean diameter, ym or ymA
E    = efficiency, fraction or %
G    = gas rate, Kg/hr-m2 column area
L    = liquid rate, Kg/hr-m2 column area
n    = particle number concentration, no./cm3
Pt   = penetration, fraction or %
Pt,  = penetration for particle diameter "d ", fraction or
Ft   = overall penetration, fraction or %
AP   = pressure drop, cm W.C. or atm
q    = vapor condensed per particle, g
q    = vapor condensed per unit mass of inlet particles,
       mass fraction
q'   = condensation ratio, g vapor condensed/g dry gas .
R    = particle concentration ratio defined in eq. (4-7),
       dimensionless
u    = velocity, cm/sec
r    = radius, cm, ym, or ymA
v-   = cumulative volume concentration, cm3/cm3
V    = gas volume swept per volume of spray,  m3/!
a    = geometric standard deviation, dimensionless
p    = density, Kg/m3 or g/cm3        •  •

 Subscripts

 a    =  aerodynamic
 d    =  drop
 i    =  inlet
 o    =  outlet
 p    =  particle
 t    =  total
                            xii

-------
                    ACKNOWLEDGEMENTS

      Air Pollution Technology, Inc., wishes to express
its 'appreciation to Dr. Leslie E.  Sparks, E.P.A., Project
Officer, and Mr. James Abbott, E.P.A. for'excellent tech-
nical coordination and for very helpful assistance in sup-
port of our technical effort.
                            Xlll

-------
                       CHAPTER 1
                      INTRODUCTION

      A major drawback of present day scrubbers is the large
 energy expenditure required to achieve high removal effi-
 ciencies for fine particles in the size range of 0.1 to 2
 microns in diameter.  This is due to the decreased effec-
 tiveness of the inertial and diffusional collection mecha-
 nisms for particles  in this size range.  Flux force and
 water vapor condensation effects have the potential to im-
 prove fine particle  collection in low energy scrubbers.
      In this report, flux forces are defined as those caused
 by thermophoresis and diffusiophoresis (which includes'the
 diffusiophoretic and Stefan flow forces); but not electro-
 phoresis.  Accordingly, we consider only those FF/C scrub-
 bers where particle  removal from the gas is aided by tempera-
 ture or vapor concentration 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.
      Several.studies of scrubber operation where particle
 collection was enhanced by vapor condensation have been
 reported.  Some investigations of the FF/C phenomena in par-
 ticulate scrubbers have been made but the results have been
 either of qualitative nature or provided limited quantitative
 information applicable only to specific cases.  Nothing ade-
 quate for the design of optimum industrial scale FF/C scrub-
 bing system was found in the literature.
      A systematic developmental study of FF/C scrubbing
 was started at Air Pollution Technology, Inc., under a pre-
 vious contract, No.  68-02-0256, where the technical and
•economic feasibilities of applying FF/C scrubbing for fine
 particle collection were established.  This study included

-------
theoretical development of design equations for FF/C scrub-
bers.  A limited bench scale experimental study was also per-
formed to examine critical areas of application of these
equations.  The. bench scale experimental st.udy was extended
further under Contract .No. 68-02-0285 to evaluate multiple
stage FF/C scrubbing, as reported in Calvert and Jhaveri ,
(1974).  It was concluded that multiple stage ,or continuous
contact type scrubbers were most suitable for FF/C scrub-
bing application.                                        .
      The purpose of the present study was to evaluate  .
technical and economic feasibilities of FF/C scrubbing
through an experimental study of a laboratory pilot, scale
FF/C scrubber.  Based on the available information,, -a
horizontal spray FF/C scrubber was selected.  'It was also -
important to evaluate the .effects of scrubber operating
parameters so that the region of optimum EE/C scrubber
operation could be defined.  To .establish the economic
feasibility, the operating costs were compared with high .'<
energy scrubbers and a case study was made to compare ;the
economics of FF/C scrubbing system against high, energy
alternatives designed to control gray iron cupola emissions.

-------
                       CHAPTER  2 .,
           SUMMARY,  CONCLUSIONS.AND RECOMMENDATIONS
SUMMARY
      Flux force and water vapor condensation effects have
the potential to greatly improve the collection efficiencies
of low energy scrubbers for fine particles.  The object of
the research reported here was to corroborate the limited
experimental and theoretical evidence of the feasibility
of FF/C scrubbing by conducting a detailed experimental
study of   a laboratory pilot scale (14 to 28 m3/min or
500 to 1,000 CFM) FF/C scrubber.
      The economic feasibility of FF/C scrubbing was also
evaluated during this study.  The results define the range
of emission properties for which FF/C scrubbing is deter-/
mined to be economical.  In general, FF/C scrubbing should
be considered when high removal efficiencies are desired
for fine particulate emissions; and the flue gas enthalpy
is higher than 100 Kcal/Kg or spent steam is available in
the plant.  These conditions are common for industrial com-
bustion processes, which include several major stationary
pollution sources in the United States.  The Midwest
Research Institute Report (1971)  ranks sources based on the
total tonnage of particulates emitted annually.
      Table 2-1 lists industrial sources of particulate pol-
lutants ranked among the top fifteen in the nation in the
M.R.I. Report (1971).  Emission properties of these sources
are favorably suited for the application of FF/C scrubbing.
The annual emissions listed were determined by subtracting
the amount of emissions removed in the existing control
equipment from the total emissions.  The "net control" listed
is the product of the average efficiency of control devices
and the application of the control devices for the industrial

-------
                 Table 2-1.
                      MAJOR INDUSTRIAL PARTICULATE SOURCES

                      FOR WHICH FF/C SCRUBBING IS ATTRACTIVE
NOTE
  IV.
  VI
The following information was taken from the Midwest Research Institute
Report (1971).  The source number (Roman numeral) refers to its rank in
the U.S.  as an industrial particulate pollution source.
                                                             NET
                                        ANNUAL             CONTROL   EMISSIONS
        SOURCE                                             FRACTION    MKg/yr
IRON AND STEEL
A. Sinter Plants (Sintering
                   process)
B. Coke Manufacture
   1. By-Product
   2. Pushing § Quenching
C. Blast Furnace
D. Steel Furnaces
   1. Open Hearth
   2. Basic Oxygen
   3. Electric Arc
E. Scarfing

FOREST PRODUCTS
A. Wigwam Burners
B. Pulp Mills
   1. Kraft Process
      a. Recovery Furnace
      b. Lime Kilns
      c. Dissolving Tanks
   2. Sulfite Process
      (Recovery Furnace)
   3. NSSC Process
      a. Recovery Furnace
      b. Fluid-Bed Reactor
   4. Bark Boilers
                                      46,300,000 MKg of Sinter

                                      81,600,000 MKg of Coal
                                      82,800,000 MKg of Coal
                                      80,600,000 MKg of Iron

                                      59,700,000 MKg of Steel
                                      43,500,000 MKg of Steel
                                      15,200,000 MKg of Steel
                                     118,800,000 MKg of Steel
                                      24,900,000 MKg of Waste

                                      22,000,000 MKg of Pulp
                                       2,300,000 MKg of Pulp
                                         756,000 MKg of Pulp
                                       3,200,000 MKg of Pulp
                                       1,100,000 MKg of Pulp
                                         470,000 MKg of Pulp
0.90

  0

 .99

 .40
 .99
 .78
 .68
  91
  94
  30

  91

 ,91
 ,70
 46,300

 81,600
 19,000
 52,600

306,000
  9,000
 16,300
 57,200
          120,000
149,000
 29,900
 38,100

  9,000

    900
 38,100
 74,400

-------
                              TABLE 2-1 (Continued)
               SOURCE

 VII.  LIME
       A. Rotary Kilns
       B. Vertical Kilns

VIII.  PRIMARY NONFERROUS METALS
       A. Aluminum
          1. Calcining of Hydroxide
          2. Reduction Cells
             a. H. S. Soderberg
             b. V. S. Soderberg
             c. Prebake

       B. Copper
          1. Roasting
          2. Reverb. Furnace
          3. Converters
       C. Zinc
          1. Roasting
             a. Fluid-Bed
             b. Ropp, multi-hearth
          2. Sintering
          3. Distillation

       D. Lead
          1. Sintering
          2. Blast Furnace
          3. Dross Reverb. Furnace

  XI.  ASPHALT
       A. Paving Material
          1. Dryers
          2. Secondary Sources
          ANNUAL
        PRODUCTION
 14,700,000 MKg of Lime
  1,600,000 MKg of Lime
  5,300,000 MKg of Aluminum

    730,000 MKg of Aluminum
    640,000 MKg of Aluminum
  1,600,000 MKg of Aluminum


    520,000 MKg of Copper
  1~,300,000 MKg of Copper
  1,300,000 MKg of Copper
    690,000 MKg of Zinc
    138,000 MKg of Zinc
    560,000 MKg of Zinc
    560,000 MKg of Zinc


    420,000 MKg of Lead
    420,000 MKg of Lead
    420,000 MKg of Lead
228,000,000 MKg of Material
   NET
 CONTROL
FRACTION
  0.81
   .39
    90

    40
    64
    64


    85
    81
    81
    98
    85
    95
    86
    83
    50
                               .96
                               .96
EMISSIONS
 MKg/yr
 267,000
   3,600
  52,600

  31,700
   9,000
  18,100
   6,000
  25,400
  29,900
  13,600
   3,600
   2,700
  13,600


  15,400
   9 ,000
   1,800
            150,000
             36,300

-------
               SOURCE

       B. Roofing Material
          1.  Blowing
          2.  Saturator

 XII.  FERROALLOYS
       A. Blast Furnace
       B. Electric Furnace

XIII.  IRON FOUNDRY
       A. Furnaces

 XIV.  SECONDARY NONFERROUS METALS
       A. Copper
          1.  Material Preparation
             a. Wire Burning
             b. Sweating Furnaces
             c. Blast Furnaces
          2.  Smelting §  Refining
       B. Aluminum
          1.  Sweating Furnaces
          2.  Refining Furnaces
         '3.  Chlorine Fluxing
       C. Lead
          1.  Pot Furnaces
          2.  Blast Furnaces
          3.  Reverb. Furnaces
       D. Zinc
          1.  Sweating Furnaces
             a. Metallic Scrap
             b. Residual Scrap
          2.  Distillation Furnace
TABLE 2-1 (Continued)

                ANNUAL
              PRODUCTION

        5,680,000 MKg of Asphalt
   NET
 CONTROL
FRACTION
                                    0.50
          540,000 MKg of Ferroalloy  .99
        1,900,000 MKg of Ferroalloy  .40
       16,000,000 MKg of Hot Metal   .27
          270,000 MKg Insulated Wire  0
           58,000 MKg of Scrap       .19
          260,000 MKg of Scrap       .68
        1,100,000 MKg of Scrap       .57

          450,000 MKg of Scrap       .19
          920,000 MKg of Scrap       .57
          120,000 MKg of CL Used     .25

           48,000 MKg of Scrap       .90
          108,000 MKg of Scrap       .90
          500,000 MKg of Scrap       .90
           47,000 MKg of Scrap        .19
          190,000 MKg of Scrap        .19
          210,000 MKg Zn Recovered    .57
EMISSIONS
 MKg/yr
              2,700
             12,700
                900
            140,000
             95,300
             37,200

              1,800
             15,400

              5,400
                900
             46,300
                900
              2,700
              2,700
              1,800

-------
category.  The information and ranks were taken from the M.R.I.
report (1971).=  It is clear from Table 2-1 that FF/C scrubbing
is a feasible and attractive participate control method for
several major industrial sources.     •                   :
Experimental Program
     Based on results from our previous studies, spray FF/C
scrubbing configurations were selected for the experimental
study.  Test aerosol with titanium dioxide particles were gen-
erated in the laboratory by dry dispersing the xrespective pig-
ment powders.  Gas heating and steam were used to precondition
the scrubber gas to the desired experimental conditions.  A
force draft cooling tower was used to cool the recirculating
scrubber water.                   ..•-'•
     Effects of several operating parameters, were studied during
the experiment.  In general, particle collection in the FF/C
scrubbers was higher as the amount of vapor condensed per unit
of gas scrubbed (condensation ratio) was increased or the inlet
particle number concentration was decreased.
     Operating parameters besides the condensation ratio and in-
let particle number concentrations which were studied for' the
spray FF/C scrubber are:  The effect of the spray drop size,    •'
the effect of decreasing cooling water requirement, the mode of
overall gas-liquid contact and the scrubber inlet liquid flow-
rate.  The scrubber performance was better when the size of   ;
sprayed drops was smaller and when the amount of cold water
sprayed in was higher for the same condensation ratio'.  The per-
formance was also better when the scrubber inlet gas was exposed
to the colder water spray first, resulting in the maximum'tem-
perature and vapor pressure gradients.
     A mathematical model for the FF/C spray scrubber based on
the 'unit mechanism approach gave only partial agreement with
the experiment.  Performance for cold gas without FF/C phenomena
benefits was compared with a mathematical model based on accoun^
ting for collection by single drops as they decelerate after
leaving the spray nozzle.  The correlation was good for particles

-------
of about 1.0 umA diameter.  However, experimental efficiencies
were higher for smaller particles!, possibly because of particle
collection on the back of drops, j Experimental efficiencies
were lower for the collection of larger particles, possibly
indicating lower utilization of sprayed liquid, higher drop co-
alescence and gas channeling.  Design equations were .developed
empirically to represent the experimental results, and to be
used for scale-up and design of industrial FF/C scrubbing sys-
tems with similar spray configuration.
Costs
     Operating costs of a FF/C scrubber were compared to those
of a high energy scrubber capable of equal performance.  The
costs of electrical power at $0.03 KWH, purchased steam at $6.40/
MKg and recirculating cooling water at $0.9/MKg were used for
this comparison.  The results indicate that when high fine par-
ticle collection efficiencies are required or when the gas to
be scrubbed is hot or humid, FF/C scrubbing is economically
more attractive than high energy scrubbers.
     In an earlier report, Calvert et al. (1973) evaluated the
economic feasibility of FF/C scrubbing compared to other alter-
natives for two industrial sources.  Based on the additional
information on the performance of FF/C scrubbing obtained from
this research, the economics of an FF/C system for a gray iron
cupola were estimated.  Good cost and performance data for a
high energy (H.E.) scrubber on a cupola were available from
another study and provided a basis for comparison.  Capital in-
vestment requirements for the two systems are roughly the same
but the H.E. sc-rubber costs about 70% more than the FF/C system
for all annual operation expenses except labor, maintenance,
liquid treatment, and solid waste disposal.  Most of the cost
advantage of FF/C scrubbing is due to a $63,500/yr higher power
cost for H.E.  scrubbing.
CONCLUSIONS
     The following conclusions can be drawn from the results of
this study:
                              8

-------
The experimental study confirmed previous predictions
of the technical feasibility of FF/C scrubbing.  High
collection efficiencies (>95%) for fine particles can
be achieved with a condensation ratio of about 0.15 g
vapor condensed/g dry gas, in a FF/C scrubber.
Of the several mechanisms involved in FF/C scrubbers,
diffusiophoresis and inertial impaction enhancement
by particle growth, which are practically independent
of particle size, are the most important for fine par-
ticle collection.  Thus, the condensation ratio and
inlet particle number are significant operating para-
meters for FF/C scrubbing.
Effects of other operating parameters, such as the gas
and liquid flowrates and contact scheme, on the perfor-
mance of the FF/C scrubber have been experimentally
studied.  The results can be used to design the optimum
FF/C scrubbing system depending on the specific proper-
ties of the industrial pollution sources.
Based on the experimental data, mathematical models
and empirical design equations' are described and can
be used to scale-up similar FF/C scrubbers for indus-
trial applications.
Based on economic considerations, the most favorable
situations for the application of FF/C scrubbing are:
(i) enthalpy of vaporization is available from the gas
to be cleaned, (ii) high collection efficiencies are
required for fine particles, and (iii) when future ca-
pacity expansion is anticipated. The smaller the size
of the particles to be scrubbed, the more economically
attractive FF/C scrubbing will look.  These conclusions
are illustrated by a case study of an economic compar-
ison of FF/C and high energy scrubbing systems for a
gray iron cupola.

-------
RECOMMENDATIONS
     The laboratory pilot scale experimental study reported here
has established the technical and economic feasibility of FF/C
scrubbing.  Engineering design equations for the spray scrubber
are also described.  The following research and developmental
program is recommended:
     1. Experimentally demonstrate the feasibility of FF/C
        scrubbing on selected industrial sources.   Pilot scale
        systems with capacities to scrub from 2.4  to 4.8 m3/sec
        (5,000 to 10,000 CFM) should be used for demonstration.
        Problem industrial sources with fine particulate emis-
        sions and hot or 'humid flue gas should be  selected for
        the demonstration.  The FF/C scrubbing systems should
        be designed to operate at the optimum conditions with
        flexibility to account for process changes.  These de-
        monstration programs,• for three industrial sources:
        a glass furnace, a secondary non-ferrous metal recovery
        furnace, a foundry cupola, are detailed in the report.
     2. Theoretical and experimental evaluation of other scrub-
        bing configurations, such as a mobile bed  (TCA)  scrubber
        should be made to determine the best configuration appli-
        cable to FF/C scrubbing systems.
     3. Evaporative cooling of scrubber liquid containing sus-
        pended and dissolved solids is critical for the  econo-
        mic feasibility of FF/C scrubbing.  The cooling  towers
        generally available use packing which are  susceptible
        to scaling by the solids.  A different configuration,
        such as using sprays instead of the packings, may be
        more applicable.  Experimental development of the
        cooling towers on a laboratory and subsequently  on an
        industrial pilot scale is clearly warranted.
     4. Theoretical and experimental study of the  spray  utili-
        zation, heat and mass transfer in gas-liquid systems,
        the nucleation of condensation and other matters which
                              10

-------
significantly influence FF/C mechanisms should be made in order
to resolve the present areas of uncertainty.
                              11

-------
                      CHAPTER 3
                      BACKGROUND

     Flux force effects on particles have been known for
many years and the background is reviewed and discussed
in depth by authors such as Waldmann and Schmitt  (1966),
Goldsmith and May (1966) , Hidy and Brock (1970) and Calvert
et al. (1972).  The studies reported by these authors in-
clude both theoretical and experimental work.  The experi-
mental systems were designed so as to be readily definable
and were much simpler than a scrubber in terms of the num-
ber of phenomena and the unsteady conditions involved.
     Several studies of scrubber operation have also been
reported where particle collection was enhanced by vapor
condensation, such as Schauer (1951) and Lapple and Kamack
(1955).  However, systematic studies of the effects of
FF/C phenomena were attempted only recently.  Rozen and
Kostin (1967) studied the collection of fine oil mist in
a perforated plate column under controlled conditions of
vapor condensation.  They found that their results could
be represented by an empirical equation relating the par-
ticle penetration, "Pt", with the mass of steam condensed
per gram of inlet particles, "q " as:

                   Pt = 12.5 qr°-56                   (3-1)

A study of steam injection into a laboratory scale scrubber
was carried out by Lancaster and Strauss (1971) .  They
measured an increase in particle collection efficiency
which was in direct proportion to the amount of steam in-
jected rather than the amount of vapor condensed.  They
concluded that the increase in collection efficiency was
due to particle growth.
     Calvert et al.  (1973)  presents a detailed description
                            12

-------
of the previously reported studies on FF/C scrubbing.  Tech-
nical feasibility of FF/C scrubbing was established in this
report, based on theoretical development of FF/C scrubber
performance models and limited bench scale experimental work.
Further bench scale experimental work, reported by Calvert
and Jhaveri (1974) led to the following conclusions:
     1. Diffusiophoresis and inertial impaction enhanced,by
        particle growth are the most significant particle col-
        lection mechanisms in FF/C scrubbers, while thermo-
        phoresis has a minor effect.  All of these mechanisms
        are practically independent of the particle size.
     2. Performance of an FF/C scrubber depends heavily on
        the amount of vapor available for condensation and the
        number concentration of particles.
     3. Multiple stage or continuous contact type of scrubbers
        are most suitable for FF/C application.  They can be
        readily adapted to provide different conditions and
        geometry along the gas path to accommodate changing
        flowrates and particle concentrations.
     4. The most critical assumptions for the application of
        FF/C scrubber engineering design equations are the
        specific details of heat and mass transfer, mode of
        gas-liquid contact and nucleation of condensation on
        the particles.  For a higher ratio of gas phase mass
        transfer coefficient to the liquid phase heat transfer
        coefficient, better particle collection would generally
        be obtained.
     Based on the above information, a spray scrubber configura-
tion was selected for experimental study.  The spray FF/C scrub-
ber has the benefit of a lower energy requirement and lower
capital costs than other low energy scrubbers because a higher
gas velocity is permissible with the spray scrubber.
     The following parameters were determined to affect the
important particle collection mechanisms (diffusiophoresis and
inertial impaction enhanced by particle growth) most signifi-

                               13

-------
cantly:
     a. Condensation ratio
     b. Particle number concentration
     c. Spray drop diameter
     d. Cold liquid flowrate
     e. Overall gas-liquid contact mode
     f. L/G
     Effects of these parameters on scrubber performance were
experimentally evaluated in this study.
                              14

-------
                           CHAPTER 4
                   EXPERIMENTAL PILOT PLANT
                                                  *
     Based on the theoretical considerations described in the
previous section, a horizontal spray scrubber was selected for
the study of FF/C scrubbing on a pilot-plant scale.  During -th^s
study, fractional penetrations of fine particles through the
scrubber were experimentally determined for several scrubber
operating conditions.  In this section, the following informa-
tion is described for the FF/C scrubber system:
     1.  Details of the pilot-scale scrubber system.
     2.  Scrubber operating procedures.
     3.  Particulate sampling procedures.
     4.  Methods of data analyses and calculation.
     5.  Accuracy of measurements.
     6.  Operational conditions studied.
THE SPRAY FF/C SCRUBBER        .     '. ;.;
Pilot Scale Scrubber System             '     :
     The schematic process flow diagram.of the FF/C scrubber sys-
tem is shown in Figure 4-1.  Components of the scrubber system
are listed in Table 4-1.  The scrubber had a design capacity of
0.47 actual m3/s (1,000.' ACFM) inlet gas flowrate.  As an- illusj
tration, flowrates in the lines shown in Figure 4-1 are'described
in Table 4-2 when the inlet stream to the scrubber is at 77°C,
saturated with water vapor.  The  FF/C scrubber and the particle
generator are described below.                          -   - ,
FF/C Scrubber                                   .
Scrubber shell:   7.6.2 cm diameter x 3.8 m long made from "Tech-
     ite" fiber glass reinforced  plastic sewer pipe, with re-
     movable end flanges.
Spray arrangement:  The scrubber  consisted of three sections,
     with a spray nozzle manifold in each section.  Each mani-
     fold had a capacity to spray 1.3 £/s  of liquid with a
                               15

-------
                     — air, scrubber gas
                     — water,  slurry
                     -— steam
                     	 natural gas
Figure -4-1.   Process flow sheet for the -FF/C spray scrubbing system.

-------
              Table  4-1.   EQUIPMENT  SPECIFICATIONS

A.  AIR PREFILTER
       (four automobile, air filters with a capacity of 8.5 m3/min air each)
B.  AIR HEATER                                                        ;
       (gas-fired air heater to heat air from 24°C to 100°C.
C.  "ABSOLUTE" AIR FILTER
       (MSA 'Ultra-Air' filter with a capacity of 0.476 m3/s of air)
D.  AIR, PARTICULATES AND STEAM MIXER
       (disc and dohut'type mixer, 0.25 s residence time) .
E.  PARTICLE GENERATOR
       (to generate dry dispersed solid particulates with a mass mean diameter
        less than 2 umA,  up to 109 particles/cm3 of scrubber inlet air)
F.  FF/G SCRUBBER
       (horizontal spray scrubber with a mesh type mist eliminator)
G.  SCRUBBER EXHAUST FAN
       (centrifugal fan with design capacity of 0.476 :m3/s and .AP of 50  cm W0C.)
H.  PUMP         ...                        ...
       (227 5,/min and 15 m head; 80 GPM and 50 ft head)
I,J,K.  PUMPS
       (114 £/min and 3.4 atm head each;  30 GPM.and 50 psi each)
L.  WATER COOLING TOWER
     ,• (a splash type, forced draft cooling tower; cooling range 47°C at
        190 i/min)
M.  BOILER WATER TREATMENT
       (11.3 £/min ion exchanger)
N.  STEAM BOILER
       (760 kg/hr steam at 2.0 atm)              .
0.  COOLING TOWER FAN                                                   .      '
       (4.25 m3/s and 2.5 cm W.C. head)                                  .
P,Q.  INTERSTAGE LIQUOR HOLDING TANKS
       (200 i capacity each)
R.  SCRUBBER LIQUOR HOLDING TANK                   ,         .
       (450 H capacity)                          ,  .           .'    .   -
SCRUBBER LIQUOR CIRCUIT                               :        -.      -
    3.8 cm nominal diameter pipes, fittings, connections and valves
STEAM CIRCUIT
    5 cm, 3.8 cm and 2.5 cm diameter pipes, connectors, valves,  steam traps;
    15.2 cm mist eliminator— all insulated
SCRUBBER GAS DUCTING
    15 cm diameter inlet duct with connectors, insulated. 15 cm diameter out-
    let duct with connectors, insulated from scrubber outlet to sampling ports

                                    17        ...

-------
          Table 4-2.   STREAM FLOW RATES OF THE  SPRAY  SCRUBBING  SYSTEM
Stream
No..
1§2
3 § 4
6
7
8
9
10
11" .
12
13
14
15
16
17
18
19
Compositions
0.024 mole H~0/mole dry air,
air mixture
n ii
0.696 mole H-O/mole dry air,
air mixture
0.0483 mole H?0/mole dry air,
air mixture
0.00675 mole H20/mole dry air,
air mixture
Natural gas
0.154 mole H20/mole dry air,
air mixture
Steam
City water
*
Natural gas
City water
0.00675 mole H-O/mole dry air,
air mixture
Process water
M it :. •'-<'
ii it
it it •
Temp .
°C
29
100
77
32
29
29
232
106
24
29
24
29
51
20
20
20
-Vol. Flow
m3/sec
0.258
0.306
0.480
0.264
0.112
0.009
0.208
-
• -
. -
-
4.75
. -
•' -
-
.
A/sec
-
-
•-
'
- -
-
-
-
0.142

-
-
3.95
1.20
1.20
1.20
Mass Flow
Kg/hr
1,028
1,028
1,450
1,060
442.8
71.2
514
••" 442
509
9 . 6
26.8
19,800
13,975
•4,525
4,525
4,525
X
Enthalpy
Kcal/Kg
9.65
42.5
206
26.8
9.65

190
644
24 '
-
24
9.65
51
20
20
20
*Air or Water

-------
     sauter mean drop diameter of 400 ym.  The nozzles were   .
   . located in a grid pattern on the manifolds to obtain a
     uniform distribution of spray in the scrubber cross-
     section.  The liquid was sprayed co-current in the direc-
     tion of gas flow.  Thirty-two size 1/4B3 "Whirljet" noz-^
     zles supplied by Spraying Systems Company were used on.
     each manifold.
Liquid flow system:  The system was.specially designed so that
     the cold scrubber liquid from the cooling tower could be  .
     sprayed either through all of the three manifolds in the  .
     scrubber ("single stage"), or through either the first or
     the last manifold in the scrubber.  In the last cases, the
     scrubber was operated as a three-stage scrubber by spraying
     the outlet liquid collected in one section of the scrubber,
     into the next section, and then into the third section in
     series.  Thus, the cold liquid requirement was reduced to
     one-third of the amount when cold liquid was sprayed in all
     of the three sections.  When the cold liquid was sprayed
     into the first section, the gas-liquid contact was co-
     current through the three sections and when it was sprayed
     in the last section, the contact was counter-current.
     Separation of sprayed liquid from one section to another
     was affected by sets of baffles, storage tanks, and pumps
     as shown in Figure 4-1.
Entrainment separator:  A wire mesh entraihment separator, 38.1
     cm diameter, 15.2 cm long, made from 0.28 mm stainless
     steel wire in a standard knit design with 98.2% voidage,
     was installed downstream of the third spray section.
Particle Generator
     The test aerosol was generated by dispersing titanium di-
oxide pigment powders.  The pigment was "Unitane OR-600", pur-
chased from the American Cyanamid Company.  The powder was
sieved through a 16-mesh screen, fed by a screw feeder arrange-
ment into a compressed air ejector, immediately downstream of a
                              19

-------
0.64 cm orifice.  The compressed air pressure upstream of the
orifice was maintained at 1.0 atm, gage.  . ,         ..
     The feed rates of the powders were controlled, depending
on the particle loading required.  A cyclone and a multiple
round jet impactor, both with cut diameters of 4 ymA, were used
in series to remove coarse particles from the dispersed aerosols.
Electrostatic charges on the dispersed particulates were neu-
tralized with nine "Staticmaster" ionizing units supplied by
Nuclear Products Company, with 500 microcuries of Polonium 210
in each.  These units were located in the .aerosol duct upstream
of the mixing section.
Instrumentation and Calibration
   .:  The inlet gas flowrate was measured with a standard pitot
tube located downstream of the air prefilter.  Scrubber system
liquor flowrates were measured by venturimeters and rotameters
installed in the 3.8 cm piping.  They were calibrated by mea-
suring the weight of water flowing through the pipes in a given
time, in the range of 0.5 to 1 H/s.  Error in the flowrate mea-
surements was kept at less than ±0.2% of the flowrate by measur-
ing at least 140 kg of water for each calibration point.
     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 poten-
tiometric amplifier.  The thermocouples were calibrated against
a standard mercury bulb thermometer using constant temperature
baths and were found to correlate well with the standard E.M.F.-
temperature relationships.  Thus, the limits of error in measur-
ing temperatures were ±0.8°C based on the manufacturer's speci-
fications.
     The scrubber gas line pressures were measured by U-tube,
inclined and well-type manometers and "Magnehelic" pressure
gauges.  Pressure drops across the liquor venturimeters were
measured with two-fluid (water and mercury)  well-type manometers.
     Moisture content in the inlet gas stream to the scrubber
was measured by wet and dry bulb thermometers.
                              20

-------
PARTICULATE SAMPLING SYSTEM
     Two identical particle sampling trains were used to measure
the particulate loadings and size distributions in the scrubber
gas inlet and outlet streams simultaneously.  The particulate
properties were measured with cascade impactors, followed by
Gelman type A glass fiber filters.  The cascade impactors used
were six and eight stage, non-viable Andersen samplers and two
University of Washington, Mark IIIF in-stack impactors.  The
samples were collected on aluminum substrates coated with high
vacuum silicone grease to prevent particle bounce.  The particu-
late loadings were rechecked by sampling with the glass fiber
filters only for each operating condition.
     The sampling instruments, either the cascade impactors or
the filter holders, were installed .in the inlet and outlet ducts
to minimize losses in the sampling probe.  Sampling probe was a
1.3 cm diameter tube.  1.3 cm aluminum tubes were installed
across the sampling instruments so that the sample flows could
bypass the instruments in each train.  The sample flowpaths were
controlled with 3-way valves.   Moisture from the sample gas was
removed by three cold impingers and a silica gel dryer located
dpwnstream of the sampling instrument in each train.
     The sampling probes, the sampling instruments and the alum-
inum tubing to the cold impingers were heated with insulated
heating tapes controlled with variacs, to prevent water conden-
sation in the lines.   The dry gas sample flowrates were measured
with a dry gas meter and a rotameter in each train.  The sampling
rates were controlled by valves on the high pressure and bypass
lines of the oil-less vacuum pumps.  Prior to each run, the sam-
pling instruments were heated to the stack gas temperatures. The
sampling lines were also heated and flushed by drawing gas through
the bypass lines.

EXPERIMENTAL PROCEDURE
     The auxiliary equipment and the measurement instrumentation
were checked and started up prior to the start-up of the scrub-

                              21

-------
her.  The liquor tanks were filled with city water, and the boiler
and cooling tower operations were started.  The scrubber was
started up by first starting the water flowrate and adjusting it
to the experimental condition and then starting the scrubber
blower to draw filtered ambient air through the scrubber.  After
a steady state was reached, the gas heater was fired and the air
was heated up to the desired temperature.  Then steam was intro-
duced to attain the experimental moisture content in the gas
stream.  Again, after a steady state was reached, the particles
were introduced to attain the experimental moisture content in
the gas stream.  Again, after a steady state was reached, the
particles were introduced to attain the experimental operating
condition.  It normally took 60 to 90 minutes to attain the ex-
perimental condition.
     The flowrates, temperatures, pressures, moisture contents
and scrubber pressure drops were measured every time a steady
state was attained during the above procedure.  These parameters
were also measured just before the start-up of particulate sam-
pling, at least once during the sampling or every thirty minutes,
and just after the sampling was completed.  Sampling time in the
scrubber inlet duct varied from 15 minutes to 60 minutes depend-
ing on the particulate loading and .the sampling rate ranged from
0.68 to 1.0 m3/hr at the probe.  The outlet sampling time was
from 2 to 3 times the inlet and the sampling rate was about twice
the inlet sampling rate to allow for the lower outlet particle
loadings.
     When the total sampling time exceeded 30 minutes, the sam-
pling was interrupted and the coarse particles caught in the cy-
clone and the impactor on the particle generator were cleaned
out.  This was necessary to maintain a steady performance of the
particle generator.  As the inlet and outlet ducts were sampled
simultaneously, small variances in the particle generator perfor-
mance did not significantly affect the results.  Also, since at
least 901 of the particles were smaller than 2 ym, the sampling
rates were not adjusted to get isokinetic velocities at the
probes.
                              22

-------
     The experimental conditions were found to stay very stable
once a steady state was reached.  For all the experimental runs
reported, the temperature conditions for the experiment varied
within ±1.5°C during the experimental period.

EXPERIMENTAL CONDITIONS STUDIED
     During the experimental study, the scrubber performance was
studied for five scrubber operational modes.  The experimental
runs for one of the operational modes, with cold water sprayed
in all of the three sections of the scrubber, were repeated to
check the reproducibility of results.  The operating modes are
described in Table 4-3.   During each mode, scrubber performance
was studied for a range of "q"', varying from the cold runs to
q'=0.2.

METHODS OF ANALYSIS AND CALCULATION
     As mentioned earlier, the particle characteristics and the
scrubber performance were measured by sampling with absolute
glass fiber filters and cascade impactors.  Information on total
particulate loadings and thus the overall scrubber penetrations,
Ptf, was obtained from both of the above sampling apparatus.  Sam-
pling with the cascade impactors provided additional information
on the particle size distributions, fractional loadings and thus
the fractional penetrations, and the inlet particle number con-
centrations.
Particle Loadings and Overall Penetrations
     The total particle loadings in the inlet and outlet ducts
were calculated in the following manner:
     1.  The sample flowrate 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 calcu-
         lated from:
                              23

-------
     Table 4-3.  SPRAY SCRUBBER OPERATION MODES

      Dust used:  Titanium dioxide, -16 mesh
                  ni = 106/cm3
NO..
1
2
3
4
5
6

OPERATING
MODE
One Stage
One Stage
One Stage
One Stage
Three Stage
Co-Current
Three Stage
Counter-
Current
INLET LIQUID
FLOWRATE PER
SECTION U/s)
-1.0
-1.0
-0.76
-0.76
-1.0
-1.0

DROP SIZE
Small
(Mass mean
dia*350 ym)
Small
Small
Large
(Mass mean
diass450 ym)
Small
Small

NOTES:  1. No.  2 was a repetition of No.  1.

        2. Liquid flow rate for No. 3 was reduced
           by reducing the number of spray nozzles.

        3. Liquid flowrate for No. 4 was  reduced
           by reducing the spray pressure.
                         24

-------
            	[	(Total weight gain,  g)              f4-i<)
           P  (Sampling rate,DNm3/min)x(Sampling  time,min)

      4.   The overall penetration was calculated  from:

                               Pt = J£°.                      (4-2)
                                    cpi                •
          where  "c   " and "cp^M were the outlet and inlet
          particle  loadings  measured simultaneously for the
          run.
 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 mo.le,-
cules.  Aerodynamic  diameter is related to geometric diameter
(actual size) by the  following,relationship:

                      dpa = dp  (C> V°'5     .          '    " (4-3>

where  d   = aerodynamic diameter, ymA  = ym
        d  = actual  diameter,  ym
        C1 = Cunningham  slip factor
        p  = particle density, g/cm3

     The Cunningham  slip factor is a complex function  of  the mean
free path of the gas and the particle diameter.  It increases as
temperature increases and as pressure decreases.   For  air at
standard temperature and pressure  it is given approximately by:

                       C' = 1 + °-165   ,        .    ,        (4-4)
                                   P
                              25

-------
     The particles were assumed to have a log normal distribution.
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 calculated from the sampling rate, based on
the manufacturer's specifications.  The geometric mass mean aero-
dynamic diameter, d  , and the geometric standard deviation, ag,
of the size distribution were then determined from a plot of the
aerodynamic cut diameters against the percent cumulative loading
smaller than the cut diameters, on a log probability paper.
Particle Number Concentration
     Particle size distribution as measured by cascade impactors
was in terms of mass and it was necessary to convert this into
a number distribution.  For log-normal size distributions it is
simple to determine number mean diameter from mass mean diameter
but this does not provide information as to the number concen-
tration.  A method was developed for computing the number con-
centration by multiplying a hypothetical number concentration
based on "d  " by a correction factor which is dependent on "ag".
           P o                                                 o
The derivation of this relationship is given below.  '
     The total number concentration of particles, n^, can be com-
puted from:
                      2H dv = nt, no./cm3           .        (4-5)
                      dv              •
and,
              dri _    5    ,  number of particles             T4-61
              dv   4 TT r 3    cm3 of particles

where  n = cumulative particle number concentration, smaller
           than "r ", no./cm3         .      •
      v  = cumulative particle volume concentration, cm3/cm3
      vt = total particle volume concentration, cm3/cm3
      rp = particle radius,  cm
                              26

-------
     We define V = — = cumulative volume fraction, dimensionless.
And,
/
                        A  -   -             dv
                        dv
     The influence of particle radius can be related to the mass
mean radius by:
                           7dn\     ,   V3
                                     r_Pg
                                    I r
                                      'P1-

where  R -  = ratio of actual number concentration for particle
             radius, r ., to hypothetical number concentration
             based on "r_ ", dimensionless
       Subscript "i" is for radius "Tp^"
       Subscript "m" is for mass mean radius "rnp"
                                               r o         '
Thus, Rn, total number concentration ratio, is:
            -i
                            actual number concentration      ,.
              R .  dV = R_ =
           ,    ni       n   no. concentration based on r_        J
         0                                             Pg
This equation was integrated numerically for different values of
"a ".  Note that for a log-normal distribution the probability
  &
of the ratio of size to mean size is normally distributed. There-
fore, "RN" is defined by "a " only.  A plot of "RN" versus "a "
was thus obtained.  From the total pa.rticulate loading and "dpg"
at scrubber inlet, the hypothetical number concentration was cal-
culated for each run.  Particle density of the titanium dioxide
was assumed .to be 3.0 g/cm3.  Then, using the experimental "^g">
n^, the particle number concentration at the scrubber inlet, was
determined from the plot.
     These values of "n^" were checked periodically by using the
Gardner and Pollack Condensation Nuclei Counters.  The, values
were found to agree within a factor of 3.  .    -       .
                              27

-------
Fractional Penetrations
     The computation of penetration as a function of particle
aerodynamic diameter, or the fractional 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" = -L
               c
             /
Pt,
dc.
                                     pa
(4-10)
where "crit" is the total particle loading and "Ptj  " is the
        ru                                        pa
penetration for particle diameter, d  , and it is given by:
                                    pa
                      fid
              Pt,

            )o
r s i
Hdpa)J
o
                                  dc
                                  (dpa)
                                            (4-11)
where
         dc
is the slope of cumulative mass loading less than
       .d(V)J
"d a" versus the aerodynamic particle diameter curve at "d'1,
and equals "f(dna)M.         ,                 . '       ;
             \ Pa/
     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 "dDa",
         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 of
         0.4 to 5 ymA.  The fractional penetrations we're then
         determined and plotted from the ratio of the slopes,
         as described above.
                              28

-------
ACCURACY OF MEASUREMENT
     Accuracy in measuring the particle size distributions, frac-
tional penetrations, overall penetrations and the inlet particle
number concentrations from the cascade impactor data depended on
several factors.  The precision of the balance, impactor handling
procedures, and measurement of sampling rate influence the sub-
sequent determination of cut diameters.  Subjective judgments of
the persons analyzing the data are involved when reading the
graphs and determining slopes.  It is beyond the scope of this
study to determine the accuracy statistically.  The best possible
care was taken in the laboratory, sampling and analytical proce-
dures to obtain accurate data and results.  At least two runs
were made at every .operating condition to duplicate the data.
     During the determination of overall penetrations using "ab-
solute" glass fiber filters, at least 5 mg of sample were col-
lected on each filter.  Precision of the analytical balance was
±0.05 mg.  Thus, the maximum error due to the weighing accuracy

               —      —   CP0
in determining Pt was  Pt = 	 <  Thus:
                      J -T-* .      P —      P *
                      arr = 	2.  _ 	i.                   r4-121
                       	                                   V-H -1- *• J
                       Pt     cp       cp.
                              O        1

     As the absolute values of the error were small compared to
the actual weights and as the error terms are additive:'
                     _              •
      Maximum error, Pt =	  +	                     (4-13)
                        _  , 0.2
                              29

-------
                          CHAPTER 5
             EXPERIMENTAL RESULTS AND DISCUSSIONS

     .Experimental and sampling procedures, and the methods of
data analyses and calculation of results are described in the
preceding chapter.  During the experimental study, scrubber
performances were 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 frac-
tional penetrations provide a common base for comparing scrubber
performances for different operating conditions.  The scrubber
operating conditions and performance are tabulated, with the
fractional penetration plots for cascade impactor runs, in
Appendix 1.
RESULTS                                                       .
     The spray scrubber was evaluated at five operating jnodes,
as listed in Table 4-3.  For each operating mode, the scrubber
performance was measured for a range of "q1" values.  Effects
of the following variables were also determined:
     1. The amount of cold water sprayed.in the scrubber was
        varied.  Effect of reducing the cold water spray in a
        3:1 ratio was studied by operating the scrubber in. a
        three-stage mode.                             •     .  .
     2. The L/G ratio was controlled by changing the inlet
        flowrate.                  f
     3. Liquid drop size distribution was changed at the same
        L/G ratio.
     4. Overall gas-liquid contact mode was altered by operating
        the three stage scrubber in the co-current or counter-
        current mode.                                '
                              30

-------
     As particle growth and diffusiophoresis are expected  to
contribute the most to the enhancement of particle capture  in
the scrubber, the effects of "q"' and "n." on "Pt" were  deter-
mined. "Pt" for 1.0 ymA and O.d^ymA particles are plotted  against
"q'M in Figures 5-1 through 5-6 for the operating, modes  studied.
Data points on these plots were read from the fractional penetra-
tion plots in Appendix 1.  The data points were found  to be,
                                                 A          . '
scattered, possibly indicating the effects of other parameters,
such as "n-" and the inlet liquid temperatures.
DISCUSSION
     The spray scrubber performance is compared with the results
of Lancaster and Strauss (1971).: and .with some plate scrubber data
on Figure 5-7, a plot of particle penetration versus condensation
or injection ratio, for'l;'0 ymA particles.  The following  is
surmised from the plo,t:r
   '•V             ,-
     1. Particle.penetration depends heavily on the condensation
        ratio, "q"1.  Calvert et al. (1973) have shown that "q"'
        is sufficient to define particle deposition rate, with-
        out regard to "n.", if there is no condensation on  the
        particles.
     2. A comparison of curves 1 and 2 shows that-"the  scrubber
                                  *•     •       s. j,'   • .>
      ,;  performance could be reproduced, if operated at nearly
        the same conditions, thus reinforcing the .validity  of
        experimental data \ai\d .technique's ....'  l...,V •1.,,.-Y'":
     3. A comparison of curvess 1 or 2 and 3 indicates  an appa-
        rent anomaly.  The amount of water sprayed into the
        scrubber during runs indicated by curve 3 was  about
        25% less than that sprayed for the runs indicated by
        curves 1 and 2.  Thus, curve 3 may. be expected to have
        higher penetrations.  This effect may be due to better
        liquid utilization when less water was sprayed in,
        resulting in a higher collection efficiency of large
        particles by impaction.  (Note that particle growth is
                               31

-------
§
I—I
E-
U
OS

E-
W
z
w
u
tx
      1.0
0.5
     0.1
    0.05
    0.01
                                             1.0  ymA
       O dpa = 1.0 ymA



       Ad   =0.6 ymA
          pa

      Refer to Table  1.1 in Appendix
     	I	I   I  I  i  i I  I I	i
       0.01
                    0.05
0.1
0.5
                           q',. g/g



    Figure 5-1 - Penetration  versus  condensation ratio

                 one stage  spray.
                         32

-------
H
U
o
t—I
H
H
W
2
w
dl
      1.0
      0.5
       0.1
      0.05
      0.01
                                        1.0  ymA
  Ad   =0.6 ymA
   pa

O d   =1.0 ymA
   pa               -        :

*Re£er to Table 1.2 in"Appendix

        .   ....  . .1     .1  .  I
         0.01
            0.05
0.1
                                                    0.5
                          '   q', g/g

      Figure 5-2  - Penetration versus condensation ratio,
                    one stage spray.
                          33

-------
    0.3
    0.2
E-
u
    0.1
o
I—I
H



H
W
2

S 0.05

w

u
a,
  0.02
                          O
         Refer to Table 1.3  ,  low L
         in
      0.03    0.05
  0.1


q1 ,g/g
0.'2   0.3
 Figure 5.3   - Penetration  versus  condensation

               ratio,  one stage spray.
                    34

-------
 1.0
2
O
i—i
E-

Crf .


<5 I
i—i

erf
w
w

w '••
u
 0.5
 0.2
 0.1
0.05
     r
     ^>
A
O
             pa
             W
                    =' 0.6 ymA
                    = 1.0 ymA
                         A
                                        0.6  ymA
        O
Refer to Table  1.4  ,  low L, large d,
in Appendix                         d
           ...... I       .     I
                0  -
    0.015
Figure 5-4
    0.03
                           0.05

                           Q', g/g
0.1
0.2    0.3
   -Penetration  versus  condensation ratio,
    one stage  spray.
                        35

-------
  1.0
  0.5
o
(-H
H
CJ
2
O
Ho.!
W
2
jjO.05
^0.03
                              1' I     '   r^ ,1
                                     ymA
O d

A d.
pa

pa
                   1.0
                    1.0  ymA
          Refer  to  Table  1.5 in  Appendix
                              1,11
      0.01
  Figure 5-5
              0.05   0.1

             q1,  g/g
                                   0.5
    -  Penetration versus 'condensation ratio,

      three  stage co-current  spray.
                      36

-------
   0.5
O
i—i
H
U
<
«;
P-.
   0.2
W
2
W
a,
u
   0.1
/
t
                       O
                           1.0
                                         A
   0.04
              O  d .  =  1.0
                  pa
     A
                     =  0.6 ymA
     Refer to Table 1.6  in Appendix

      .     !    .  ,""," .  .  I       i
                                         A
                                         O
       0.02
   Figure  5-6
      0.05        0.1        0.2

             q',. g/g

-  Penetration versus condensation ratio,

  three stage counter-current spray.
                                               0.4
                         37

-------
  "1.0
   0.5
o
hH
H
U
<
(*
PH
E-
W

W
(X

W

u
I—I
E-
Qi

O,
   0.1
   0.05
  0.01
         — — — —Lancaster § Strauss  (1971)
          Numbers refer to Table numbers
          in Appendix
   ... 0.01..      .   .   .0.05   ,Q.l-     .  	  0.5
               CONDENSATION OR INJECTION RATIO'

                  (g vapor/ g dry gas)

  Figure  5.7  -  Comparison of FF/C plate and spray
                 scrubber results for 1 ymA particles
                       38

-------
   higher at higher "a"' valjues.)  Spray drop diameters
   during all the runs were -the same.  The better liquid
   utilization may be due to1 less drop coalescens.e and
   lesser wall losses at the lower liquid flowrate.  This
   effect'bears further experimental investigation.
4. A comparison of curves 3 and 4 illustrates the effect
   of spray drop diameter.  Although the liquid and gas
   flowrates were identical for these runs, the spray drop
   volume mean diameter for curve 4 was about 1-5 times
   the diameter for curve 3.  Thus, the particle collec-
   tion by impaction and the spray drop space density (num-
   ber of drops in a unit scrubber volume) are lower for
   curve 4, resulting in the high penetrations.
5. A comparison of curves 1 or 2 (1 stage) with 5 (co-
   current), shows that the particle penetrations are
   higher when the amount of cold liquid sprayed is,reduced
   by about 67% although the total liquid spray rate is
   the same.  In both the operating modes cold liquid was
   sprayed in the section where the gas enthalpy and vapor
   content were highest.  Thus, the highest possible tem-
   perature and vapor concentration gradients were imposed
   in this section.  The slightly higher penetrations for
   curve 5 indicate that although most of the particle
   growth occurred in the first section of the scrubber,
   FF/C mechanisms were also effective in the second and
   third sections.
6. A comparison of curves 5 with 6 (counter-current) indi-
   cates the importance of causing particle growth quickly
   in a FF/C scrubber.  The only difference in scrubber
   operation was that for curve 5, the coldest water was
   sprayed into the gas with the highest enthalpy and water
   vapor content; while for curve 6, the hottest water was
   sprayed into this gas in the first section, thus pro-
   viding a more uniform distribution of gradients along
                          39

-------
the length of the scrubber.  Curve 5 penetrations are
lower, probably due to more particle growth for co-
current contacting.
                      40

-------
                           CHAPTER 6
         FF/C SCRUBBER PERFORMANCE PREDICTION METHODS

     The prediction of particle collection performance for FF/C
scrubbers in advance of experiments can be done by means of
theoretically based mathematical models.  The approach used is
to derive particle collection equations which account for all
of the applicable unit mechanisms, which could be active in the
scrubber.  Such models are then used to predict particle collec-
tion in the spray FF/C scrubbers.  The predicted performances
are compared to the experimental results.
     Particle collection in the FF/C sp-ray scrubber is affected
by inertial impaction, diffusiophoresis, and thermoporesis.
During this study our approach was to develop equations for
particle collection by inertial impaction, and use them in con-
junction with.the mathematical models for particle growth and
collection by flux forces as described by Calvert et al. (1973).
     Particle collection efficiency by inertial impaction in
a spray scrubber can be predicted by means of methods described
in the "Scrubber Handbook" (Calvert et al. 1972), for cases :
where the drop velocity is constant, or may be considered so.
When the spray is generated by high pressure nozzles, the drop
velocity is initially very high compared to the terminal settling
velocity.  Therefore the collection efficiency of the drop de-
creases greatly as the drop slows down and the overall collection
by the drop is the integrated effect of efficiency over the
drop trajectory.
     -Walton and Woolcock (1960) studied this problem in connec-
tion with the use of pressure sprays to control coal mine dust.
They computed the relationship between collection efficiency
and the distance traveled by a drop for several particle, sizes
and drop diameters.  Figure 6-1 is taken from their paper and
shows these relationships as predicted for coal dust (density =
1.37 g/cm3).  Drop velocity is also plotted so that one can find

                               41

-------
- ii.SL
                                         Velocity,  tn/sc-c    V. .  in/sec
                                       .ill  :n   III   5 J II     --=	
                     lull

              Dis.tancc,  cm
H 20       SO    '    I)  Jil -10

  !>i s t ,-ii:cc , cm       Distance, cm
                                                                                  100
                                                                                    10
                                                                                                                      L
                                                                                     "0        2 .S         5.0        7.5        10

                                                                                      HORIZONTAL niSTAMCH FROM SPRAY  MANIFOLD,cm

                                                                                      l;igure 6-2 --Scrubber area covered  by sprays.
            !-"iEiire li-l  -  I'vcdictcil pnrticlc collection  efficiency
                 • .    •   for  sprays by inertial  impaction  and interception

-------
efficiency as the drop accelerates from a given initial velocity.
     The relationships given in Figure 6-1 can .be used to predict
the collection efficiency of a spray scrubber making :the following
assumptions which we used in our model:
     1. Assume that the particle concentration is uniform as
        the aerosol enters a spray stage, that it decreases
        exponentially as it passes through the stage, and that
        it is completely mixed (i.e. uniform concentration)
        before entering the next stage.
     2. Assume that collection is by inertial impaction on the
        front of the drop only.  This is the same as Walton
        and Woolcock assumed.
     3. Assume that the percent of the gas area covered by
        the sprays (co-current), varies as shown in Figure ,6-2,
        which is based on the arrangement of nozzles used in
        our pilot plant spray scrubber.
     4. Assume that the drop diameter is uniform.
     'Based on the above assumptions, the equations, describing
the multi-stage, co-current spray scrubber are .as follows:
     Volume of gas which is. swept clean of particles per unit
of liquid volume is:    •       . "  .       :               ~   .
           Vs =  (^f-) RdE x id-'-3, (ni3/*)              (6-1)
where
     V  = gas volume swept clean per liter of drops, m3/^
     r_, = drop -radius, cm
     R, = drop range (i.e., distance traveled), cm
     E  = average particle collection efficiency over range
          "Rd", fraction         • • .   •  •    -  -  •

     If the collection of particles is a first order process,

                     PtA = exp - '(Vs ± )             ,   (6-2)
                               43

-------
where L/G = Liquid to gas flow ratio,  O H20/m3 gas)
      Pt. = Penetration per stage for a given particle
            diameter, fraction

     The penetration for "N" stages is:

          .  PtN = (PtA)N  .                          (6-3)

     The average efficiency was computed by plotting the
product of efficiency times fraction of gas flow covered
by sprays, versus the drop range (distance traveled) and
then doing a graphical integration.  These plots were
made for several particle sizes and for an initial velo-
city of 20 m/sec and drop diameters of 0.05 cm and 0.03
cm.  These conditions correspond to our pilot plant runs
at 2.7 atm (40  psig) spray nozzle pressure.  We also
assumed that the maximum drop range is 100 cm, based on
scrubber size.
     The results of the computations for a liquid to gas
ratio of 2.35 £/m3 (18 gal/MCF), corresponding to the
flow rate per stage in our pilot plant runs, and for
three stages are shown in Figure 6-3 for 0.05 cm drops
and Figure 6-4 for 0.03 cm drops.  Because of the differ-
ence in slope between the predictions and the experimental
data we explored the influence of decreasing the amount
of effective spray, assuming that there is agglomeration
The results are shown in Figures 6-3 and 6-4.
     As can be seen, the predictions are fairly close to
the data at the cut diameter (0.5 Pt) but not for other
particle diameters. We have not yet been able to devise
a model that will account for this discrepancy.  It is
quite possible that lower penetration for particles smaller
than 1.0 ymA is due to their collection on the backs of
the drops.  However, we do not yet have a predictive cor-
                           44

-------
-L . U

a
0 0.5
g 0.4
a;
. 0.3
a
o
H
EH 0.2
a;
•En
W
w '
04
0.1
_ . 1 I 1 1 - 1 « 1
|
—
123 •,'-..-
- \\
\\
-_ \
- Curve L/G,£/m3 ',
1
1 2.35 \
-2 1.0 .
3 0.7
i i i i i i 1
:i
\

\\.
. i
\ \
\\
\ \
\\
1
	
—
'_


'

         0.3    0.5       1.0     2.0.

              PARTICLE DIAMETER,
Figure-6-3.   Spray scrubber penetration predictions
              for 500 ym drop diameter.
     1.0
   a
   o
   H
   EH
   £
a
o
H
EH

«
EH
W
a
   0.5


   0.4
     0. 3
     0.2
     0.1
1 1 111 1 1
- 12
1 1
3
_
—
; \\ \
\



- Curve L/G,^l/m3

1 2.35
' ' 2 1.0
3 0,7
i i.ii.l
. \\
\ \\
\ \\
\ v\ .
\ \ \
\ \ ^
\ \\
1


_

•^
• -

        0.3
Figure 6-4.
            0.5       1.0       2.0

           PARTICLE DIAMETER, ymA

           Spray scrubber penetration  predictions

           for 300 ym drop diameter.
                   45

-------
relation to account for this effect.  Since a design
equation for particle collection by inertial impaction
could not be validated by our data, a model ;for FF/C spray
scrubber was not developed.
     In order to obtain design equations for scale-up pur-
poses of FF/C spray scrubbers similar to the scrubber tested
in this study, a curve fitting technique was used.  Scrubber
performance data for 1.0 ymA particles are plotted in Fig-
ures 6-5 through 6-7\ as particle penetration versus the
amount of vapor condensed per particle, "q".  This value
was used instead of "q"'as it normalizes the effect of
particle number concentration, "n.".  Again, scatter in
the data was observed, possibly due to the effects of inlet
liquid temperatures and the actual amount of vapor condensed
on the particles.  The range of scatter around the least
square fit was within  1501.
     Curve fitting procedures were employed on each hori-
zontal spray FF/C scrubber data set to obtain the functional
relationship between "Pt" and "q".  The initial step is the
assumption that the relationship between "Pt" and "q" can
be represented as a power function:

          Pt = AqB                                   (6-4)

or synonymously as the straight line equation:

          In Pt = In A + B In q                      (6-5)

     The method of least squares was utilized to obtain the
best straight line curve fit through the experimental set
of paired variables, "In Pt" and "In q", where "In q" was
regarded as an ordinary variable measured without appre-
ciable error, and "In Pt" as the random variable.
                             46

-------
  a
  o
  H
  EH
  U n.l
   •O.oi
        o see  table  1.1
          see  table  1.2
          (in  appendix)
                  i
       -12
                  -11
                               -10
      10"'-        in"11         in

      q'  g vapor condensed per particle
                                                       i .n
                     2
                     O
                     H
                     EH
                     U  "-1
                                                      0.01
                                                           o see table
                            Asee  table 1
                             (in  appendix)
                                	i
                                                          -12
                                                                     -11
                          in Li        in"11         in"1"         in •

                          q1  g vapor condensed per  particle
Figure 6-5.   Experimental  1.0 ymA particle    Figure  6-6.  Experimental 1.0  ymA particle
               penetration for  spray  scrubber.                penetration for  spray  scrubber,
                            i.o.
                          O
                          H
                          EH
                          U
                          -P
                          ft
                           •n. o i
 o see  table  1.6

 A see  table  1.5
	(in  appendix)
                              10
                               -12
                                                     10
                                                      -10
                                                                10
                            q1  g vapor  condensed per particle

                       Figure  6-7.  Experimental 1.0 ymA particle
                                     penetration  for spray scrubber,

-------
     After the determination of the constants "A" and "B"
for the straight line equation, the percent error for each
data point was calculated.  Those that indicated appre-
ciable error (251 or greater) were eliminated (these points
are shaded on the plots) and the least square straight
lines were redetermined for the remaining number of data
points.  As shown in Table 6-1, four data points were thus
not included in the determination of the least square
straight lines out of a total of fifty-two points.  This
resulted in narrower bands for confidence limits and thus
precluded the accounting of a typical scrubber performance.
     The 90% confidence intervals for the mean values, "Pt",
of the regression curve:

          In Pt = A + B In q                         (6-6)

were established utilizing the t-distribution table, 90%
confidence level, and n-2 degrees of freedom where "n"
was the number of data points.  The equations are presented
in Table 6-1 and by straight lines on Figures 6-5 through
6-7.   The 90% confidence intervals are also represented
by broken lines on these figures.
     In lieu of a rigorous mathematical model,, the equations
developed above can be used for predicting particle collec-
tion in FF/C spray scrubbers.  For particles smaller than
2 ymA in diameter, the collection is mainly dependent on
particle growth and flux forces.  Thus the penetration was
expressed as a function of "q" in the equations.  The equa-
tions can be used to predict the amount of vapor condensa-
tion required in the scrubber for the designed penetration
of fine particles.  Application of the above equations is
limited to the design of FF/C spray scrubbers with the con-
figuration and operating parameters similar to the pilot
                            43

-------
                   Table 6-1.  FF/C SPRAY SCRUBBER DESIGN EQUATIONS
        SCRUBBER OPERATING MODE
One stage mode, cold water sprayed in all sections
L/G ~ 2.5 £/m3 each stage
  DESIGN EQUATION
             - 5  -0.3 1
Pt = 7.5 x 10  q
Single stage mode, cold water sprayed in all
sections, L/G x 2.1 £/m3 each stage, small
liquid drops
             -7 -0.5 3
Pt = 2.7 x 10  q
Single stage mode, cold water sprayed in all
secitons, L/G =2.1 £/m3 each stage, large
liquid drops
             - 5 - O.C 6
Pt = 3.4 x 10  q
Three stages with co-current "liquid/gas contact
scheme, L/G ~ 2.5 &/m3 each stage
Pt = 1.5 x 10"2q"°'°
Three stages with counter current liquid/
gas contact scheme, L/G z 2.5 £/m3 each stage
Pt = 1.7 x 10~V°'2

-------
scrubber evaluated in this study.  It is noted however
that the pilot scrubber was operated at near optimum
conditions for maximizing particle collection and thus
covers the practical design range for the horizontal,
FF/C spray, scrubbers.
                            50

-------
                          CHAPTER 7  :                      . .  ;.
                    • ECONOMIC FEASIBILITY  '

     Experimental results discussed in the previous sections, ' -
together with the theoretical and experimental studies reported
by previous investigators, clearly show that fine particles can
be collected with high efficiency in FF/C scrubbers.  The eco-
nomic feasibility of FF/C scrubbing is discussed in the follow-
ing section.  Since there has been no published study of an
industrial FF/C scrubber, actual data on the economics of such
a system are not available.  Thus, the discussion is limited
to preliminary predictions of costs based on the available in-
formation on FF/C scrubbing.
     Some general economic features of FF/C scrubbing are dis-
cussed below.  Experimental results plotted on Figure 5-7 in-
dicate that it should require from 0.1 to about 0.25 g water va-
por condensed/g dry gas in a FF/C scrubber to attain high col-
lection efficiency for fine particles.  Such a condensation ratio
generally requires preconditioning of the scrubber inlet gas to
increase its moisture content.
     Gas preconditioning could be done either by direct intro-
duction of spent steam if the gas is dry and has low enthalpy,
or by the evaporation of sprayed water when enough enthalpy,
is available in the gas.  Direct injection of.steam is beneficial
because it can increase the local saturation ratio above 1.0,
which is necessary for the growth of hydrophobic particles.
     Cooling water is needed to condense the desired amount
of vapor in the scrubber.  In an industria.1 system the water.
is cooled in an evaporative cooling tower using ambient air,
and then, recirculated to the FF/C scrubber.  In cooling towers
of conventional design, the cooling range is kept below about
17°C'(30°F).  A larger water temperature range can be achieved
in cooling towers of special design but the costs will be higher
                               51

-------
than usual, and there may be undesirable features such as fog
formation.  If the water temperature rise in the scrubber is
17°C (30°F), about 32 g of cooling water will be required to
condense 1 g of water vapor.
COST COMPARISON
     It is likely that one would have to make a choice between
using high pressure drop or FF/C conditions in a scrubber system
for fine particle collection.  If equipment costs for the two
types of system are roughly the same, most of the difference in
operating costs will be due to power, water, and steam require-
ments.   In order to compare the two approaches, operating costs
have been estimated for an FF/C system and a high energy (venturi)
scrubber and the results are described below.
     As an example case we have taken flowrate of 1,700 Kg/min
of dry gas (D.G.) with molecular weight of 29.0 and an initial
humidity of 0.01 g H20/g D.G.  Various inlet gas temperatures
are considered and it is assumed that the gas will reach its
adiabatic saturation temperature in the high energy (H.E.)  scrub-
ber and 49°C in the FF/C scrubber outlet.  A 10°C lower outlet
temperature than the assumed 49°C from an FF/C scrubber could
generally be attained without great difficulty so that 49°C
assumption is conservative.  The saturated gas is assumed to
travel from the scrubber to an induced draft fan and then to
discharge.  Thus the fan power requirement will depend on the
humidity, temperature, and pressure of the scrubbed gas.
     Costs'were estimated for several operating modes of an
FF/C scrubber and for some combinations of inlet gas temperature
and pressure drop for an H.E. scrubber.  Some illustrative re-
sults are shown in Figure 7-1, a plot of hourly operating costs
against condensation ratio with parameters of scrubber type and
operating conditions.
     Assumptions and cost bases for Figure 7-1 are as follows:
     1. Inlet dry gas flowrate is 1,700 Kg/min, pressure is 1.0
atm abs., humidity is 0.01 g H20/g D.G., molecular weight is
                              52

-------
    25
    20
A
^
•be-
    15
    ib
L •



^
w

&  10
         iff
        r.:.t
               -id!
               tit
               11--!
                     rf+r
                              -4+
                            rr::±r irrtt"
                  &"  &
                  •rrtr.
                               iitr.
                                     -tffi
                                  rrr.r
                               } . i.
                               !:«::
                                      Tit
                                         -Lt:
                                            tn
                                            rft
                                               Hit

                                               fflt-
                                               rrT"
                                                 I+t
   i
"p.: T-
-it:.'.'
                                                   :r±fr
                                                              ::±t
                                                            V+]
                                                            •H-H-
                                                              1
                                                                    "Tf
                                                  K
                                                     .rft
                                                             ±r
                                                              w
                                                                      •tt
                                                                      .i-..
                                                                     Trrt
                                                                  ±!:T. .,..
                                                                  •^t-H+iili-
                                                                  -T-H-.i-i-rt-
                                                                  •t-rT-t--r»-t4
                                                                  -.-t-r -r r-- f
                    0.1         0.2         0.3         0.4

                       CONDENSATION  RATIO  (q!),  g/g


            Figure  7-1.   Operating cost  comparison of
                            FF/C  and  H.E.  scrubbers.
                                  53

-------
29.0.  Various inlet temperatures are considered.
     2. Cooling water could cost from 0.26
-------
                       Table  7-1.   GAS  CONDITIONS AND FAN POWER COSTS
Inlet Gas
Temp.
C°C)
260
400
500
800
1000
Humidity
(g/gD.G.)
0.01
0.01
0.01
0.01
0.01
Ad. Sat. Gas
Temp.
(°C)
49
58
66
71
75
Humidity
(g/gD.G.)
0.08
0.14
0.22
0.3
0.39
Fan Power Cost for 1,700 Kg D.G. /min, ($/hr)
Pressure Drop, cm W.C.
15cm
$2.67


--

25cm
$4.47
--

--

48cm
$ 9.06
9.80
11.20
12.40
13.50
60cm
$11.50
12.40
14.00
15.60
17.00
90cm
$17.70
19.30
21.80
24.20
26.00
en
l/i
              Table  7-2.   TYPICAL PENETRATIONS FOR PLATES AND SPRAY

Pilot Scrubber
5 Sieve Plates
Sprays
Penetration for 1 ymA dia. particles at q' shown
0.05 g/g
0.22
0.13
0.1
0.17
.0..092
0.15
0.12
0.075
•0.2
0.078
0,057
0.25
0.06
0.04

-------
typical values of experimental penetration for 1 umA diameter
particle as a function of condensation ratio for plate and spray
type FF/C scrubbers so that the cost data can be readily inter-
preted.
     The three dashed lines are for the fan power costs associated
with H.E. scrubbers operating at 48, 60, and 90 cm W.C. pressure
drop (as indicated on the plot) and at actual volumetric flow-
rates corresponding to a given condensation ratio.  To illustrate
the meaning of these lines, gas at 600°C and humidity of 0.01
g/g would reach an adiabatic saturation temperature of 66°C and
humidity of 0.22 g/g in a H.E. scrubber.  The volumetric flow-
rate due to 1,700 Kg/min of dry gas and 374 Kg/min of water vapor
would be about 2.2 x 103 m3/min at 66°C and 1.0 atm absolute
pressure.  The volumetric flowrate will be higher at lower abso-
lute pressure corresponding to negative inlet fan pressures equi-
valent to the pressure drop across the scrubber system.
     The cost of fan power is computed from the actual volumetric
flowrate and pressure drop by the use of the relevant assumptions
and costs given above.  In order to compare the cost for H.E.
scrubbing to that for FF/C scrubbing, they have been plotted
against the condensation ratio which could have resulted if gas
at the inlet conditions had been treated in an FF/C system.  For
the 600°C inlet temperature the condensation ratio would be the
difference between the adiabatic saturation humidity and the
assumed FF/C outlet humidity; that is, q'=0.22-0.08 = 0.14 g/g.
     At q'=0.14, Figure 7-1 shows the following:

              Scrubber     Pt @ 1 ymA    Cost. $/hr

            FF/C Spray        0.08          8.80
            H.E. 48 cm W.C.   0.1      ,    11.00
            H.E. 60 cm        0.05         13.70
            H.E. 90 cm        0.02         21.50
            H.E. 42 cm*       0.125         9.50

   Note: *Computed for comparison at same Pt as FF/C plate.

                               56

-------
     It can be seen from the above data that FF/C scrubbing would
require lower operating costs than H.E.  The cost advantage of
FF/C scrubbing increases as the inlet gas enthalpy increases
and the penetration requirement decreases.  If low penetration
is not required a low pressure drop H.E. scrubber may give satis-
factory performance at lower cost than FF/C.  The point where
H.E. and FF/C scrubbing will have the same operating cost for
the same efficiency, depending on FF/C scrubber type, is a gas
temperature of about 400°C.  The condensation ratio would be
about 0.06 g/g and the predictions from Figure 7-1 and some addi-
tional computations for H.E. scrubbers are as follows:
              Scrubber       Pt @1  mA       Cost, $/hr
            FF/C Spray         0.12             6.30
            H.E. 30 cm W.C.    0.21             6.30.
            H.E. 44 cm         0.12             9.00 .   .
     In the "cold" operation mode (i.e. q'~0), the spray gave
the. same or slightly better efficiency for a given power input
than other types of H.E. scrubbers, such as gas atomized sprays.
For example, at about 2.4 &/m3 in each of three sections, as in
the pilot scale spray scrubber, the liquid pumping power would
be equivalent to about 17 cm W.C. pressure drop in terms of fan
power.  The cut diameter of the spray scrubber is about 0.9 ymA.
For a H.E. scrubber with a pressure drop of 17 cm W.C., the
cut diameter if 0.92 ymA.  Since the penetration at a given
"q"1 is lower for the spray than the H.E. scrubber, pre-formed
sprays will be economically superior to H.E. scrubbers where
there is any significant amount of condensation taking place.
Steam Introduction
     While the performance of an FF/C scrubber at a given con-
densation ratio is better if part or all of the water vapor is
introduced as steam (i.e. 100% HaO), the cost of purchased steam
will generally be prohibitively high.  However, under the right
circumstances the use of some steam introduction could be eco-
nomical.
                              57

-------
     The steam required for injection could be low pressure <2 atm
gage, ( <30 psig) spent steam.  It may be obtained from the feed
line to the boiler condenser in a.process plant or 'generated in
a low pressure waste boiler in a metallurgical operation.  In
this case, the steam cost would be significantly lower and will
depend on the specific manufacturing process.  In general, if
such steam is available for less than $1.88/MKg, the operating
costs for a FF/C scrubber would be lower than for a venturi
scrubber.
     Figure 5-7 indicates that the "q" requirement levels out
around q'=0.15, so that proportionately less condensation is re-
quired to obtain penetrations lower than 5% in a FF/C scrubber.
Thus, in this region, a FF/C scrubber using some purchased steam
would have lower operating costs compared to a venturi scrubber.
Also, as the most important factors in FF/C scrubbing, diffusio-
phoresis and particle growth by condensation are practically in-
sensitive to particle size,  FF/C scrubbing would become econo-
mically more attractive as the size of the particles to be con-
trolled gets smaller, in the range of 0.01 ym to 10 ym diameter.
Industrial Application Costs
     Calvert et al. (1973) have evaluated the economic feasi-
bility of FF/C scrubbing systems designed for two industrial
sources:  A Basic Oxygen Furnace and a Kraft Recovery Furnace.
The gas cleaning devices in these systems were a FF/C spray
scrubber for the Basic Oxygen Furnace and a combination of a
venturi evaporator-scrubber followed by a FF/C condenser vessel
with a spray scrubber for the Kraft Recovery Furnace.
     The economic feasibility of FF/C scrubbing for a gray
iron cupola is evaluted below.  Due to the different designs
and operating practices for cupolas, it is not possible to
generalize emission characteristics so a specific cupola was
selected for the case study.  Emissions from this cupola are now
controlled with a high energy scrubber whose performance was
measured by A.P.T. (Calvert et al. 1974).  Information on emis-
sions, system behavior and costs were obtained from the cupola

                               58

-------
 operators.   The  FF/C  scrubber  system was  designed for the same
 level of particle  control  as  the  existing system and the equip-
 ment  and operating costs are  compared.
 CUPOLA EMISSION  CONTROL
      Gray iron foundries use  cupola,  electric-arc,  electric-
 induction arid reverberatory air furnaces  to  obtain  molten metal
 for production of  castings.   The  iron melting"5'process is the
 principal source of emissions  in  the  foundry industry.   Cupola
 furnaces are most  commonly used for  the melting  operations.
 In the foundry industry, cupola emissions are dominant,  totaling
 over  105,000 ton/yr as reported by M.R.I. (1971).
      The cupola  emissions  are  presently controlled  by high energy
 scrubbers,  fabric  filters  (baghouses) and electrostatic  precipi-
 tators.   FF/C scrubbers are uniquely  suited  for  cupola emission
 control,  due to  the combination of high stack gas  temperatures
 (over 900°C) , fine particle sizes  (d_ -1.5 ymA) ,  moderately high
 dust  loadings (over 3 g/DNm3)  and the presence of contaminant
 gases.   .
      The FF/C system  described below  was  designed for a  2 m  (80")
 diameter cupola.   Presently the cupola emissions  are controlled
 with  a high  energy scrubber system.   Details on  the performance
 of this  system together with  the  cost data have  been reported  by
 Calvert  et  al. (1974).  The cupola operation is  breifly  des-
 cribed below:
 Size:   2 m  (80") I.D., 9.8 m  (32') high from tuyeres to  after-'
        burners and 7.6 cm  (25') additional height to the top.
        The  shell is water  cooled  to  a height of  6 m (20') and
        then  refractory lined  to the  top.   The fuel  gas flow to
        the  two afterburners is controlled to maintain the off-
        take  gas  temperature between  870°C and 930°C.
Charge:  Each  charge consists of about  2,700 Kg of  metal  (mostly
        case  iron),  between 270Kg  to  320Kg of coke and fluxes
        as required.   On an average there  are 12  charges  per hour.
                                59

-------
      The cupola emissions were reported to be as follows:
Flue gas rate = 3,400 Am3/min (119,500 ACFM) @ 900°C.
Flue gas humidity = 0.01    i       , from humidity of the
         ambient air.
Particulate characteristics:  3.5 g/DNm3 (8 gr/DSCF) , log
         normally 'distributed with mass mean diameter,,
         d  =1.15 ymA and a =1.7.
          fO               O
The present high energy scrubber system has an overall par-
ticulate removal efficiency of 97% at the scrubber pressure
drop of 275 cm W.C. (Calvert et al , 1974).
The FF/C Scrubber System                  •
      The FF/C scrubber system was designed to yield 971
particulate collection efficiency, identical to the per-
formance of the high energy scrubber.  A spray configura-
tion was selected for the FF/C scrubber due to the low
pressure drop requirement.  Based on the FF/C spray scrub-
ber tested in this study, q'=0.18 is considered sufficient
to achieve the desired particulate removal, with the over-
all liquid sprayed to gas flowrate ratio of 4.5 £/m3(34
gal/MCF) .
      A process diagram of the FF/C scrubber system is
shown in Figure 7-2.  As described earlier, stack gas from
cupola offtake is maintained at about 900°C by adjusting
the afterburners.  Immediately downstream of the -refrac-
tory-lined offtake, 150 £pm of water is sprayed in as fine
mist, cooling the gas to 427°C.  This serves the purpose of
reducing heat losses from the 90 m long duct to the scrub-
ber system, as well as permitting the use of smaller dia-
meter mild steel duct.  Gas temperature at the quencher in-
let is determined to be 280 °C after accounting for .the heat
losses.
                            60

-------
ENTHALPY
.LOSS  FROM  DUCT
                           TO I. D.
                          FAN, STACK
                                                    AIR
FROM
CUPOLA
WATER '
AYS 1
1 T2
t

'8
QUENCHER
>
>
^9
i
*-
i*
>
f
\
- s
.
CAUS
FOR
CONT
f





/C
RUBBER
fJ-> CJ
P-, en

TIC
nil -—
PM
ROL
'

>
7
6
r
LIQUOR'
TREATMENT





*M
1
->J
->-
••
-X

COOLING
TOWER
i ,
AIR
BLOW
Tr\
                                                        MAKE-UP
                                                        WATER
                                                 AND SLUDGE
                                                 TO  DRA-IN





CO

u


Q
1— 1
JD
cx
t— 1
H-)
STREAM
NO.


1
2
3
4
• 5..
6
7
8
9

TEMP
°C


900
427
280
69
49
27
46
64
69

VOLUME •
FLOWRATE
Am3/min
or &/min
3,400
2,480
1,850
1,300
970 .
5,870
6,040
400
300

GAS,.
HUMIDITY
g vapor
g D.G.
0.01
0.16
0.16
0.26
0.08





PRESS
cm W.C.


-2
' -7
-15
-20
-35





LOADING
g/ENm3


3.5
3.5
3.5
. 3.5
. 0.105





    FIGURE 7-2 - Process diagram for cupola gas cleaning
                           61

-------
     The gas is then further cooled and humidified by re-
circulating water sprays in the quencher.  As the size
distribution of particles in the stack gas is fine, with
over 99% by mass of particles smaller than 5 ymA, water
sprays in the quencher are not expected to remove any sig-
nificant amount of particles.  The desired particulate re-
moval is achieved in the FF/C spray scrubber.  The scrubbed
gas is cooled down to 49°C in the scrubber, condensing out
                                           ' ' ?          -
0.18 g vapor/g d gas on the liquor sprays.  The overall pres-
sure drop across the system is "determined to be- 35 cm W.C..
     The liquor system includes a cooling tower and provi-
                                             t
sions for clarification and pH control of-the recirculated
liquid.  The uniformly packed cooling tower uses ambient
air for evaporative cooling of the liquid sprayed in the
scrubber.  Clarification of the recirculated liquid is
attained by settling out-the suspended solids in tanks,
with a total retention time of about 45 minutes.  The sett-
ling process is enhanced by the addition of coagulants.
Caustic soda solution is added in the tanks to maintain the
pH between 6 and 7.  A constant blowdown from the tanks con-
trols the concentration of dissolved solids in the liquid  \
stream.  A water wash is provided in the cooling tower to
periodically clean out scale formed on the packing surface.
City water is added into the cooling tower to make up for
the blowdown, entraihment, evaporative and other losses from
the recirculated liquid system.
Cost Comparison
     Equipment for the system shown schematically in Figure
7-2 was selected and sized for the purpose of cost estima- .
tion.  Capital cost information was obtained from the fol- ',
lowing sources:  A.P.T. Scrubber Handbook (1972) , Cost   .  -,
Engineering in_ the Process Industries (I960) , Mo,de,rn Cost,
Engineering Techniques (1970) , and Chemical Engineers'
Handbook, Fifth ed.. , (1973).  The method was based on

                            62

-------
calculating the F.O.B. equipment cost and then multiplying
by various factors for the costs of internals, piping, in-
strumentation, etc.  The capital costs were then adjusted
to a common time base, 1974, using the Marshall '§ Stevens
Equipment Cost Index.
     Installation costs for the scrubber system, to include
modifications in existing process, site preparation, foun-
dation, start-up, etc. were not determined, as they are ex-
pected to be comparable to the costs;for the high energy
scrubber system.  Similarly, the operating costs for labor,
liquid treatment, and solid disposal, together with the
maintenance costs for labor and materials, are expected to
be comparable for the two scrubber systems.  Thus, the only
operating costs compared are the electrical power require-
ments and annualized capital charged and depreciation taken
as 20$ of the capital costs.  Table 7-3 shows the cost com-
parisons.
     The FF/C scrubber system requires additional equipment
such as the quencher and the cooling tower.  Due to the
higher flow requirements of recirculated liquid, the .liquid
system costs are also higher for the FF/C system compared
to the high energy (H.E.) scrubber.  However, these costs
are more than offset by the higher cost for fans for the
H.E. scrubber system.  Three fans are required for the H.E.
system, adding up to 1,080 KW (1,450 HP) as compared to one
127 KW (170 HP) fan required for the FF/C system.  Note that
costs for piping and ducting for both the systems are ex-
pected to be the same, although they are significantly dif-
ferent.  The FF/C system has a higher piping requirement
and requires the 90 m duct to be lightly insulated.  These
costs are expected to approximately offset the cost of addi-
tional ducting used in the H.E. system to jacket the 90 m
duct.
                            63

-------
Table 7-3. COST COMPARISON OF CUPOLA.EMISSION'CONTROL.SYSTEMS

COST ITEMS

A. Capital Costs
1. F.O.B. quencher with
internals, flange to flange
2. F.O.B. scrubber with
internals, flange to flange
including entrainment
separator
3. F.O.B. cooling tower
4. Fans, motors and
motor starter
5. Liquid treatment .and
solid handling equipment,
including pumps .
6. Piping and ducting (2)
7. Instrumentation and
electrical material (3)
TOTAL EQUIPMENT COST .
B. Annual Operating Costs
1. Electrical power for
fans and pumps.
2. Annualized capital
charges and depreciation
(20% of capital costs)
TOTAL
High Energy
Scrubber
System ($) (1)'
-

-- (4)'



18,600


156,370


50,030
102,570

. 22,900
350,470


102,560


70,100
$172,660
FF/C Scrubber
System ,
"c$r
'

12,140



32,460
. 34,950

35,000


80,000
102,570

15,640
: 312,760 .. ,


38,900


62,550
$ 101,450
  Notes :
   1. Actual  costs obtained  from  the user,  converted  to  1974
  3

  4
Due to equivalent complexity, the costs were assumed
same for both systems.

Taken as. 5% of equipment costs for the FF/C system..

Quench spray costs for both the systems are included
in the ducting costs.
                             64

-------
     A comparison of costs for the two systems indicates
that the total equipment costs are approximately the same.
The H.E. system, however, costs about $63,500 more per
year to operate, as the annual cost for electrical power
is more than 2.6 times that for the FF/C system.  Electri-
cal power cost of $0.03/KW-HR was used for the above cal-
culations.   Power costs have increased steadily in recent
years and there are no immediate indications for a change
in this trend.  As the cost of electrical power increases
and energy conservation becomes more important, the FF/C
scrubbing system will prove to be more attractive, compared
to a high energy scrubbing system.
                            65

-------
                          CHAPTER 8
              FUTURE RESEARCH RECOMMENDATIONS .

     The preliminary objective of examining the technical and
economical feasibilities of FF/C scrubbing for ,the collection
of fine particles, through experimental evaluation, has been
achieved in this study.  It has been clearly shown that FF/C
scrubbing is capable of high efficiency fine par.ticulate re-
moval.  Areas of economic application of FF/C scrubbing at
industrial sources have been identified.  These include some
of the major stationary air pollution sources in the U.S.A.
as listed in the Midwest Research Institute report (1971).
Mathematical design models have been developed also for the
spray scrubber studied.  To continue this development work so
that the advantages of FF/C scrubbing could be derived for
industrial application, we recommend future research work in
the following areas:

      1.   Demonstration of the feasibility of FF/C scrub-
          bing on selected industrial sources.
      2.   Theoretical and experimental evaluation of other
          low energy scrubber configurations to determine
          the best configuration applicable to FF/C scrub-
          bing systems.
      3.   Development of evaporative cooling devices suited
          for the cooling of scrubber liquid containing
          suspended and dissolved solids.
      4.   Theoretical and experimental determination of the
          specific details of heat and mass transfer in gas-
          liquid systems, the nucleation of condensation
          and other matters  which critically affect the
          applicability of FF/C scrubber design equations.
                              66

-------
DEMONSTRATION OF FF/C SCRUBBING
    A detailed test program to demonstrate FF/C scrubbers
at pilot scale for the control of fine participate emissions
from three industrial sources is described below.  The fol-
lowing criteria were used for selecting the industrial sources
      I.  The national importance of the industrial parti -
          culate emission sources as major pollutants.
     II.  Applicability of FF/C scrubbing in terms of'its
          technical and economic feasibility.
    III.  Sources which are either difficult or expensive
          to control with presently available particulate
          control devices.            .
    Three sources were selected which would enable demon-
stration of the operational reliability of all the compo-
nents of an FF/C scrubbing system.  These sources are
described below:
Secondary Nonferrous Metals Recovery Furnace
    The secondary nonferrous smelting and refining indus-
try generally uses gas-fired furnaces to recover copper,
aluminum, lead and zinc from scrap and dross.   The nature
of furnace operations is such that emissions vary widely
during the cycle from charging the scrap to pouring a melt.
Peak emission surges occur in nearly all the furnace opera-
tions.  The principal emissions from these furnaces are par-
ticulates in the form of smoke, dust, and metallic fumes.-
However, during copper wire reclamation, considerable amounts
of acidic and corrosive gases comprised of fluorine and chlo-
rine compounds are also present, depending on the composition
of the wire insulation.  With emission rates of about 127,000
ton/year, this industry was ranked among the top fifteen
national source pollutants in a survey by Midwest Research
Institute (1971).

                            67

-------
    Due to the cyclic nature of the emission, high flue
gas temperatures O5000C to 900°C), presence of corrosive
gases and high efficiencies required for fine particulates,
the recovery furnace emissions have been difficult and ex-
pensive to control.  Generally, afterburners are used fol-
lowing the furnace to burn out hydrocarbons in the flue gas.
High energy scrubbers, electrostatic precipitators and
fabric filters have then been applied to remove' the particu-
lates.  None of these have proven reliable for maintenance
free operation.  No control device is presently available to
control emissions from copper wire recovering operations,
although over 300,000 ton of insulated wire is recycled an-
nually resulting in emissions exceeding 41,000 ton/year.
    We recommend that FF/C scrubbing should be demonstra-
ted on a recovery furnace, especially one operated for copper
wire reclamation.  A pilot scale system with a capacity of
140 to 280m3/min is recommended.  A small furnace with a pro-
cess load of about 450 Kg/hr,  with 20-minute cycles, would
be ideally suited for the demonstration.  The total flue gas
emission of such a furnace would be in the range of the pilot
scale FF/C scrubber capacity so that the effect of the cyclic
nature of emission could be best evaluated.  This would^per-
mit also a study of the FF/C system performance for the simul-
taneous removal of fine particulates and corrosive gases,
using alkaline scrubber liquor.
Glass Furnace
    The glass manufacturing industry , and  especially  the con-
tainer glass industry,  is faced with a nation-wide need for
the application of particulate control systems on the glass
furnaces.   Glass furnace emissions have been difficult and
expensive to control due to a high fraction of fine particulates,
                            68

-------
high stack gas temperatures and the presence of gaseous
contaminants.  However, these emission properties are
favorably suited for economic application of FF/C scrub-
bing.
      A typical furnace produces from 80 to 140 metric
tons of glass per day on a 24 hrs/day schedule.  Typical
flue gas properties for a furnace producing amber glass
are as follows:
      Gas Conditions:
          Flowrate:  300 m3/min @ 20°C             .  .-
          Temperature: 450°C
      Particulates:
          Loading: 0.2 g/m3 @ 20°C    -
          Emissions: 4 Kg/hr.
          Composition: ~90% NaaSCs
                       -10% CaSO,,
                       trace amounts of other constituents
      The particles are -1.0 ymA, with geometric standard
deviation less than 3.
      Gaseous contaminants:
          SO : 80 ppm
            X.
          NO : 1,500 ppm
            -A.

      In addition to the removal of gaseous contaminants,
over 801 removal of particulates is required and the opacity
limited to less than 20%.  An FF/C scrubber has the additional
advantage of particle growth, as - 90% particulates are soluble
in water and would grow at a saturation ratio of less than
one.  This would be one of the major aspects of the demon-
stration.  Again, a pilot scale system with a capacity of 140
to 280m3/min is recommended.
                           69

-------
Gray Iron Foundry Cupola
     Gray iron foundries use several types of furnaces to  '
melt and recover iron from scrap for the production of
castings.  Cupola furnaces are most prominently used for
the melting operations.  The iron foundry industry was rank-
ed among the top 15 national stationary-source pollutants
in the Midwest Research Institute (1971) survey.  In the
foundry industry, cupola particulate emissions are dominant,
totaling over 105,000 ton/year.  Physical processes^ chemi-
cal reactions, and the quality of scrap affect the emissions
of dust and fumes from cupola, thus no typical flue gas con-
dition can be defined.  For example, emissions from a 86 cm
(34") cold blast cupola with a production rate of 45 Kg
(100 Ib) molten iron per minute are listed below:
     Flue gas rate = 520 A m3/min (18,350 ACFM) § 980°C
     Flue gas rate = 115 N m3/min (4,000 SCFM)
     NOTE:  The flue gas rate and temperature are maintained
            constant by adjusting gas flowrate to an after-
            burner located above the charge door.
     Average flue gas composition, before afterburn, vol $:
            C02:  12%
            CO :  14.9%
            N2 :  73.1%
     Particulate loading:  1.0 g/DNm3  (2.3 gr/DSCF)
     The particulate size distribution data reported in the
literature for cupola emissions were found to vary consid-
erably from source to source.  Due to the carbon particles
present in the emission, the particles are considered to
be.non-wettable.
     Cupola emissions have been difficult and expensive to
control due to the high emission rates of fine particulates,
high gas temperatures and significant changes in the emission
characteristics during the operation cycle.  At present,
                            70

-------
electrostatic precipitators,  high energy scrubbers and
fabric filters (bag houses) are used to control these emis-
sions.  The high gas temperatures and fine particulate
loadings prompt the economic  use of FF/C scrubbers.  Again,
a 140 to 280m3/min pilot scale FF/C system is recommended.
     A detailed demonstration test program, including the
cost and time estimates, is described below.   Although the
details of FF/C scrubbing system design will  be different
for each source, the overall  process des.ign,  illustrated
in Figure 8-1, will be the same.  Since the test matrices
for the demonstrations would  also be of comparable complex-
ities , we expect that the cost and period of  performance for
each demonstration will be the same.  Any variations in
these estimates can be easily accommodated;when more details
on the .installation 'and operation of the industrial source
in question are available.
     In outline, the. .objectives consist of the following
tasks:
     1.  Select a company which operates- a suitable plant
         involving one of the -above operations for the
         demonstration of FF/C scrubbing and  obtain re-
         quired clearances from the local air' pollution
         control agency.
     2.  Design the demonstration scrubber system on the
         basis of:
         A.  Pertinent data concerning the source obtained
             through source testing.
         B.  Evaluations of alternative FF/C  scrubber
             system designs.
     3.  Prepare a detailed test plan describing:
         A.  The measurement  techniques to be used.
         B.  Error analysis of the measurement techniques.
         C.  The proposed test matrix.
         D.  The data handling procedures.

                            71

-------
K>
Gity

Water-
       0)

       o
       c
       •
              thW.. .Ij-tzfT
                     To drain or

                    Liquid Treatment


g
o
  V

CJX)
UH
  O

  C/3
      City

      Water
                                                    To stack
                                                       t
                                              Air
                                                     00

                                                     C (H
                                                     •H 0)

                                                     rH ?

                                                     O O

                                                     O E-

                                                     U
                                                           Air
         Figure 8-1 - Typical process design of a FF/C scrubber system

-------
     4.  Fabricate, install and start up the FF/C scrubber
         system as developed in 2 above.
     5.  Conduct the test program as developed in 3 above.
     6.  Remove the pilot FF/C scrubbing system and restore
         the site to its normal condition less fair wear and
         tear.
     7.  Conduct an engineering and cost analysis of the
        , FF/C scrubber system to evaluate:
         A.  Scrubber operating and capital cost.
         B.  Scrubber operation and maintenance problems.
         C.  Scrubber reliability.
         D.  Scrubber performance.
     8.  Based on the above analysis and the additional
         information available on scrubber design, design
         and estimate the cost of an optimum FF/C scrubber
         system for the industrial source.
     9.  Define the areas of FF/C scrubbing in which addi-
         tional information is needed to improve the via-
         bility of the FF/C scrubber application.
    10.  Survey emission conditions for similar sources to
         determine a group of industrial operations where
         the optimum FF/C scrubbing system developed above
         is economically applicable with minor modifications,
    11.  Recommend a test program to demonstrate a full-
         scale FF/C scrubbing system on a simliar industrial
         source.
     For the purpose of estimating the cost and time require-
ments, the following task breakdown was used:
   Task 1 - Select company
     a.  Survey the industry and contact companies
     b.  Preliminary screening
     c.  Contact screened candidates to .obtain additional
         information and perform sampling, where needed.
                             73

-------
  d.  Refine calculations of conceptual designs
      based on "c" above.
  e.v  Consult the contracting agency.
  f.  Select the company for the demonstration.
  g.  Consult the local air pollution control agency.
  h.  Finalize the arrangements.
Task 2 - Design the demonstration pilot plant
  a.  Obtain more data on the source
      1.  Sample the source
      2.  Conduct small-scale (10-20 CFM)  tests on
          the source to test out concepts  for scrubbing,
          liquor cooling, monitoring, etc.
  b.  Evaluate alternatives by means of:
      1.  Preliminary designs
      2.  Laboratory bench scale tests
      3.  Laboratory pilot scale tests
      4.  Additional small-scale tests at  source
      5.  Finalize alternative designs
      6.  Compare alternative designs
      7.  Consult contracting agency
      8.  Consult company
      9.  Clear with local air pollution control agency
     10.  Make detailed design and specifications of
          demonstration pilot plant.
Task 3 - Prepare detailed test plan
  a.  Develop plan
  b.  Error analysis
  c.  Define procedures
  d.  Finalize measurement and monitoring  procedures
Task 4 - Fabricate, install, start up
  a.  Select vendors and place orders            , .
  b.  Arrange for site modifications
  c.  Follow up vendors and assemble components
  d.  Install pilot plant
  e.  Start up pilot plant

                         74

-------
   Task 5 - Conduct test
     a.  At near-optimum conditions
     b.  At excursions from optimum
     c.  Configuration variations
   Task 6 - Remove pilot system and restore site
   Task 7 - Engineering analysis            '
     a.  Operating and capital costs
     b.  Operating and maintenance problems
     c.  Reliability
     d.  Performance
         1.  Scrubber
         2.  Liquor system and other auxiliaries
   Task 8 - Optimum FF/C system
     a.  For this plant condition
     b.  For other capacities.                   -
   Task 9 - Evaluate FF/C scrubbing system performance to
            define areas for additional investigation
   Task 10- Survey emissions from other sources to deter-
            mine a group of operations with similar
            emission problems
   Task 11- Recommend full scale test program
     A performance schedule for this program is described
in Table 8-1.  The overall period of performance is 18
months for the demonstration program.  It is noted that
additional demonstration programs performed on the other
industrial sources would require 14 months, if performed
by the same contractor.                         •> • ••
     A detailed cost breakdown for the demonstration program
is described in Table 8-2.  It is noted that the costs were
based on 1974 prices and for the FF/C scrubbing system only.
If there are special construction and installation problems
at the industrial source, the costs should be adjusted ac-
cordingly.  It is expected that additional demonstration
                            75

-------
TABLE 8-1: ESTIMATED SCHEDULE OF PERFORMANCE
Task
No.
la
b
c
d
e
£
£
h
2a 1
2
b 1
7
3
4
5
6
7
8
9
10--



Month #1

X
















































/
X





















X


























/


/





-










2


























y
























x

























/
/





















X



















7





X
























X

/
/
/
/
/



















X
























X
























X



/










4















/

/
/













































/


















X



























x



/




5




















/



















X
























X
X



















-



















x








x




6














/





































































































x




-------
TABLE 8-1:  (Cont.)
Task
No.
2 b 3
4
3 a
b
c
d
4 a
b
c
d
e
5 a
b
c
7 a
b
c
d





Month #7



/
/

/
/
















































/
























/





















X
















8





/





























































































X
X



















9





















































X


X

Y





























































10






































































































X


x
X













11











K






•-
























































































X












12












/



/

/














































































X
X






















-------
               TABLE 8-1:  (Cont.)
oo
Task
No.
5 a
b
c
6
7 a
b
c
d
8
9
10
11











Month #13





/













-













































































r







.














14


/




X


























































































X





















15



/





/

























































































x




















16





















































X
V
^
















































>
-------
 programs  at  the  other  industrial  sources would  cost  about
 80%  overall,  if  performed  by  the.same  organization.
             Table  8-2.   DETAILED COST BREAKDOWN
Task
No.
1
2
3
4
5
6
7
8
9
10
11
TOTAL
Direct
Labor
($)
16,000
32,000
15,000
19,000
19,500
6,500
4,500
4,500
4,000
5,000
4,000
130,000
Other
Direct Costs
($)
4,000
18,500
5,000
40,000
4,500
	
2,500
	 , .
	
	
	
74,500
     The above estimates were made on the assumption that
the performing organization is eminently qualified in the
research and development field, with special knowledge, back-
ground and interest in the specific field of particulate re-
moval by FF/C scrubbing.
                              79

-------
EVALUATION OF OTHER SCRUBBER CONFIGURATIONS

     During the performance of this contract,  the horizontal
spray scrubber was evaluated, when operated in the FF/C scrub-
bing mode.  Several other configurations of conventional and
novel designs are presently available for low  pressure drop
scrubbing, such as the mobile bed scrubber and the packed bed
scrubber with conventional and new packings.   We recommend that
these configurations be evaluated to determine their applica-
bility to FF/C scrubbing.  The following approach is recommended:

     1.   Survey scrubber  manufacturers  and users  to  assimi-
         late  design and  operating  information on low  energy
         scrubbers  with high  gas  residence times.  The scrub-
         bers  should be multi-stage  or  continuous  contact
         type.
     2.   Screen and select  promising scrubber  configurations
         based on theoretical  evaluation  of the  scrubber per-
         formance when operated  in  the  FF/C scrubbing  mode,
         for  the collection of fine  particles.
     3.   Conduct a limited  bench  scale  study to  obtain in-
         formation in the critical  areas  of scrubber opera-
         tion,  so that the  FF/C  scrubber  model could be
         theoretically derived.
     4.   Determine  the technical  and economic  feasibilities
         and  the special  area  of  application for  these scrub-
         bers  by using the  models developed above.   The
         feasibilities should  be  determined by a  comparative
         evaluation of all  the FF/C  scrubbers  considered.
     5.   Select promising scrubber mechanisms  and conduct  a
                                                     i   '
         detailed experimental study as  follows:
                                                          ~« . .
         a.   Laboratory pilot  scale  study with scrubber  capa-
             city between 14  and  28  m3/roln.
                              80

-------
       b.  Pilot scale demonstration on selected industrial
           sources with scrubber capacity between 140 to
           280 m3/min.
       c.  Full scale demonstration on a selected industrial
           source.
     It is noted that performance of each of the above, tasks
would depend on the results of the preceding tasks.   The
program objective is to determine FF/C scrubber configura-
tions best suited for specific industrial operations or a
group of operations with common particulate control  problems.

DEVELOPMENT OF LIQUOR 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 a FF/C scrubber system.  The
scope of the proposed work is to evaluate the liquor cool-
ing system alternatives, select the most promising system
and make an experimental study to determine its perfor-
mance, economics and applicability.
     The following approach is recommended:
     1.  Determine all factors affecting the liquor  cooling
         system for a FF/C scrubber.  These include  such
         things as economic considerations, effect on
         scrubber performance and the concentrations of   \
         dissolved and suspended impurities in the liquor.
     2.  Evaluate cooling system alternatives and select
         the most promising for experimental study.
     3.  Conduct an experimental study on a suitable scale
         to determine its performance, economics, appli-  .
         cability and scale-up considerations.
                             81

-------
                          REFERENCES
 Calvert, S., J. Goldshmid, D. Leith, and D. Mehta.  Scrub-
 ber Handbook.  A.P.T., Inc.  EPA Contract No. CPA-70-95.
 NTIS  #PB 213 016.  August 1972.

 Calvert, S., J. Goldshmid, D. Leith, and N. Jhaveri.
 Feasibility of Flux Force/Condensation Scrubbing for Fine
 Particulate Collection.  A.P.T., Inc.  EPA Contract No.  .
 68-02-0256.  NTIS  #PB  227 307.  October 1973.

 Calvert, S.  Engineering Design of Fine Particle Scrubbers.
 J. Air Pollution Control Association.  24  (10) :929-933.
 October 1974.                      -                   -  .

 Calvert, S., and N. C. Jhaveri.  Flux Force/Condensation
 Scrubbing.  J. Air Pollution Control Association.  24  CIO):
 947-951.  October  1974.

 Calvert, S., and S. Yung.  Evaluation of Venturi-Rod Scrub-.
 ber.  A.P.T. Inc.  EPA Contract No. 68-02-1328, Task No. 5..
 August 1974.

 Chilton, C. H.  Cost Engineering in the Process Industries.
 New York, McGraw-Hill.  P. 475.  1960.

 Goldsmith, P., and F. G. May.  Diffusiophoresis and Thermo-
 phoresis in Water Vapor Systems.  Aerosol Science.  New '
 York, Academic Press.  P. 163-194.  1966.

 Handbook of Emissions, Effluents and Control Practices for
 Stationary Particulate Pollution Source.  Midwest Research
 Institute.  Report to NAPCA, Contract No. CPA-22-69-104.
 1970.

Hidy, G.  M., and J. R.  Brock.   The Dynamics of Aerocolloidal
Systems.   New York, Pergamon Press.   P.  379.   1970.

Lancaster, B.  W., and W.  Strauss.   A Study of Steam In-
jection Into Wet Scrubbers.   Ind Eng Chem Fundamentals.
10(3) :362-369.   March 1971.

Lapple, C.  W.,  and H.  J.  Kamack.  Performance of Wet Dust
Scrubbers.   Chem Eng Prog.   51(3) :110-121.   March 1955.
                             82

-------
Perry, J.H. Chemical Engineer's Handbook, 5th ed. McGraw-
Hill, 1973.

Popper, H.  Modern Cost Engineering Techniques.  New York,
McGraw-Hill.  1970.

Rozen, A. M., and V. M. Kostin.  Collection of Finely Dis-
persed Aerosols in Plate Columns by Condensation Enlargement
Inter Chem Eng.  7:464-467.  July 1967.

Schauer, P. J.  Removal of Submicron Aerosol Particles from
Moving Gas Stream.  Ind Eng Chem.  43(7): 1532-1538.  July
1951.


Waldmann, L.., and K. H. Schmitt.;  Thermopho.resis and Dif-
fusiophoresis of Aerosols.  Aerosol Science.  New York,
Academic Press.  P. 137-161.   1966.

Walton, W. M., and A.  Woolcock.  The Suppression of Airborne
Dust by Water Spray.  Inter J Air Poll.  3:129-153.  October
1960.
                             83

-------
           \APPENDIX 1

  HORIZONTAL SPRAY FF/C SCRUBBER
OPERATING CONDITIONS AND PERFORMANCE
                 84

-------
Table  I . 1   HORIZONTAL SPRAY SCRUBBER; OPERATING CONDITIONS AND
             PERFORMANCE          .             '


Bust used:  Titanium dioxide, -16 mesh
Spray  configuration:  Cold water sprayed in all of the three
                      sections and drained out through a common
                      outlet.  Thus, scrubber operated in the
                      single stage mode.
Inlet  water flowrates:  First Section: 1.05 liters/sec @ 2.67 kg/cm2
                      Second Section: " 1.09 liters/sec @ 2.67 kg/cm2
                      Third Section:   1.0  liters/sec @ 2.85 kg/cm2
Run
No.
1
2
3
4
5
6
7
Gas Inlet' Conditions
Flow
(JNm3
'-mm ;
26.2
25.9
25.3
24.6
24.8
20.8
22.0
Temp
C°C)
14.8
12.8
44.5
45.5
44. 5
46.3
49.3
Moisture
(% vol.)
0.6
0.6
9.5
10.0
9.9
9.9
9.6
Relative
Humidity
w
29
35
92
97
100
92
78
Liquid Temp.
. C°c)
. In
11
11
18
19
19
17
16
Out
9
9
20
24
24
21
21
Scrubber Pressure
Differences Ccm W.C.)
APS
+ 0.9
+ 0.9
+ 0.9
+ 0.9
+ 0.9
+ 0.9
+ 0.8
APE
-0.7
-0.6
-1.1
-1.0
-1.0
-1.1
-0.7
AP0
+ 0.2
+ 0.2
-0.2
-0.1
-0.1
-0.2
+ 0.1
Run
No.
1
2
3
4
5
6
7
Particulate . •
. Load x 103(g/DNm3)
In
192.0
. 180.0
106.2
91.1
107.6
138.1
53.3
uut
84.0
84.0
8.3
11.9
11.1
13.3
3.5
dn.
In
1.00
1.13
1.09
0.99



Out
1.00
1.10
0.96
0.95
a
In
2.0
1.9
2.0
1.7
Filter run
Filt
er run
Filter run
i
g
Out
1.8
1.6
1.4
1.6.



ni
•xlO'6
6.8
4.4
3.2
1.0



q'
xlO
	
	
0.5
0.5
0.5
0.5
0.5
PT
%
43.8
46.7
7.8
13.1
10.3
9.6
6.6
Note:  APC :  pressure difference across the scrubber
         o                                        -       •

       APp :  pressure difference across the entralnment separator
         Xj                                ,

       AP~ :  overall pressure difference, AP~ - AP_ ,

             positive sign denotes pressure gain while pressure
             loss is denoted by a negative sign.
                              85

-------
                   Table  1.1  (continued)
00
Run
No.
8
9
10
11
12
13
14
15
Gas Inlet Conditions
Flow
(JUS)3
lmirr
22.9
22.8
23.7
23.0
22.3
21.9
22.3
23. S
Temp
CO
57.3
57.0
58.0
57.5
64.5
66.5
67.0
67.8
Moisture
(t vol.)
15.6
15.7
16.6
16.8
23.3
2S.7
25.5
25.5
Relative
Humidity
Ct)
86
86
86
90
90
90
89
85
.iquid Temp.
CO
In
20
21
21
20.5
22
22
23
23.5
Out
25
27
26
25.5
40
43
45
44.5
Scrubber Pressure
Differences (cm W.C.)
4PS
+ 0.9
• 0.8
*1.0
+ 0.9
*1.0
+ 1.0
«1.0
»1.1
4PE
-0.9
-0.9
-1.1
-0.9
-1.0
-1.0
-1.0
•-1.0
ipo
...
-0.1
-0.1
...
---
...
...
+ 0.1
                                                                                                             Table  1*1  (continued)
Run
No.
8
9
10
11
12
13
14
IS
Particulate
Load x 10
In
72.8
110.6
113.4
97.3
58.7
46.6
136.0
154.7
!(g/DNi,')
Out
5.4
5.1
5.2
3.7
2.3
2.6
7.6
7.5
dp*
In
1.13

1.31


1.36
1.29

Out
0.96
"*
In
1.7
Out
1.6
Filter run
0.92
1.9
1.6
Filter run
Filter run
1.00
0.90
2.0
1.9
1.6
1.6.
Filter run .
"i
xlO~*
2.0
'
1.5


0.7
1.9

,?;
0.9
0.9
1..0
1.0
l'.5
1.6
1.6
1'. 7
r
7.5
4.6
4.5
3.8
3.9
5.6
5.6
4.8
Run
No.
16
17
18
19
20
21
22
23
Gas Inlet Conditions
Flow
(-SSJ
'•mm'
21.1
17.2
17.7
17.7
21.5
19.8
19.2
22.2
Temp
CC)
68.5
73.5
73.0
74.0
75.5
81.5
79.5
78.0
Moisture
C* vol.)
27.7
34.1
35.0
34.0
28.8
32.7
33.6
• 42.5
Relative
Humidity
(*)
90
91
95
89
70
62
.70
95
.iquid Temp.
CO
In
22
25
27
27
22
27
25.5
24
Out
46
SO
55
52
44
53
48
49
Scrubber Pressure
Differences (cm W.C.)
ips
+ 0.9
+ 0.8
+ 1.0
»0.9
• 1.0
+ 1.0
+1.0
+ 1.0
fiPE
-0.9
-0.7
-0.7
-0.6
-0.9
-0.7
-0.8
-0.6
AP0
	
+ 0.1
+ 0.3
+ 0.3
+ 0.1
+ 0.3
+ 0.2
+ 0.4
Run
No.
16
17
18
19
20
21
22
23
Pal
Load X 10
In
147.0
77.1
102.6
75.8
105.1
73.6
79.8-
68.9
'Cg/DNm3)
Out
5.8
4.3
3.5
3.5
24.7
17.0
17.2
6.13
ticulate
"V
In
1.30
1.32

1.40

1.21
t.ll
1.45
Out
0.94
1.09
",
In
1.9
2.4
Out
1.7
1.5
Filter run
1.05
1.9
1.5
Filter run
0.98
0.97
0.98
2.0
2.2
2.2
1.7
1.7
1.6
"i
xlO's
2:0
3.4

0.8

1.6
3.7
1.3
q'
ill)
1.8
2.5
2.5
2.8
2.0
2.5
2.9
3.6
r
3.9
5.6
3.4
4.6
23.5
23.1
21.6
8.9

-------
                     Table
                              1.2
                                  HORIZONTAL SPRAY SCRUBBER; OPERATING CONDITIONS
                                  AND PERFORMANCE
CO
                     Dust used:   Titanium dioxide, -16 mesh.
                     Spray configuration:  Cold water sprayed in all of :the three
                                 sections and drained out through a common outlet.
                                 Thus, scrubber operated in the single stage mode.
                     Inlet water flow rates:  First section - 1.05 t/sec 8 2.67 Kg/cm2
                                 Second section - 1.09 -I/sec 8 2.67 Kg/cm! .
                                 Third section - 1.0 t/sec 8 2.8S Kg/cm2
Run
No.
24
25
26
27
28
29
30
31
Gas Inlet Conditions
Flow
, DNiN3
lmTnJ
23.7
23.0
23.9
24.2
24
23
22.8
23
Temp
(°C)
10
12
11.5
11
11
46
45.5
45
Moisture
(4 vol.)
0.76
1.0
1.1
1.4
1.1
10.1
10.3
10
Relative
Humidity
c»)
71
68
96
77
77
95
97
100
Liquid Temp.
(°C)
In
8
8
8
8
9
16
15.5
14.5
Out
9.5
9.5
9.5
9.5
10.5
23.5
21.5
22.5
Scrubber Pressure'
Differences (cm W.C.)
ips
0.8
0.8
0.8
0.8
0.8
0.8
0.8
0.8
APE
1.0
1.0
1.0
1.0
1.0
0.9
0.9
1.0
AP0 •
-0.2
-0.2
-0.2
-0.2
-0.2
-0.1
-0.1
-0.2
Run
Ho.
24
25
26
27
28
29
30
31

Load x 103(g/!)Nm')
In
122.4
52.3
97.4
98.4
106
90.2
104.0
100
Out
47.43
21.7
37.8
41.6
46.6 -
9.41
9.47
11.9 .
V
In

1.2

1.07
1.05
1.2

1.06
Out

1.02

1.04
0.96
1.06

1.0
",
In

1.7

1.8
1.8
1.9

1.8
"Out

1.6

1.7
1.7
. 1.6

1.7
ni
xlO-6

0.6

1.9
2.3
1.5 '

1.9
q'
xin
__
--

--
--
0.48
0.50
0.50
¥
39.
42.
38.
42.
44.
10.
7Z
12.
                                                                                                        Table   /.2 (continued)
Run
No.
32
33
34
35
36
37
38
39
40
• Gas Inlet Conditions
Flow
nNm3
'•mm'
22.1
21.9
21.7.
20.7
20.3
19.5
18.8
19.6
19.6
Temp
C°C)
56
57
56
66
65.5
82.5
73.5
84.5
84.5
Moisture
(t'vol.)
17.3 .
17.4
16.4
26.04
26.9
36.3
1.2
35.1
36
Relative
Humidity
(»)
97
90
95
95
100
66
90
60
57
.iquid Temp.
("C)
In
16
18
22
23
19
23
24
22
22
Out
28
31.5
34
38
38
49
48.8
47.5
47.5
Scrubber Pressure
Differences (cum W.C.)
ips
0.8
0.8
0.9
0.8
0.8
0.9
0.8
0.9
0.9
APE
0.8
0.8
0.9
0.8
0.8
0.8
0.8
0.7
0.7
4P0
0
0
0
0
0
0.1
0
0.2
0.2


Run

32
33
34
35
36
37
38
39
40
,.

Load x 10
In
133
128
101
121.0
158.0
131.0
128.4
146.0
144.0 .

Pal
3(g/DNnr>)
Out
6.02
9.29
8.59
6.37
22.4
10.0
' 22.1
- 22.1

ticu]
d
In
1.23
1.04
1.01
	
1.06
1.21
1.16
	

ate

Out
0.9
0.84
0.79
Filte
0.88
0.77
0.96
Filte:
9

C
-In"
1.9
1.5
1.6
Ru
1.8
1.9
1.8
Ru



Out
1.9
1.5
1.7.

1.7
1.62
1.5



ni
xlO'6
2.0
1.3
1.9

3.1
2.1
2.3




q'
xin
1.01
1.01
1.9
1.75
2.20
2.45
2.65
.2.27



\?
045
09.2
07.1
04,0
17.
OZ8
15.
IS.

-------
                  Table    1.3.  HORIZONTAL SPRAY SCRUBBER; OPERATING CONDITIONS
                               AND PERFORMANCE

                  Dust used:  Titanium dioxide, -16 mesh
                  Spray configuration:  Scrubber operated in single stage node.
                             The number of nozzles was decreased to obtain the
                             lower water flowrates, maintaining the small spray
                             drop -diameter
                  Inlet water flowrates:  0.76 I/sec in each section
Table   \ .4.  HORIZONTAL SPRAY SCRUBBER; OPERATING CONDITIONS
              AND PERFORMANCE

Dust used:   Titanium dioxide, -16 mesh
Spray configuration:  Scrubber operated in the single stage
            mode.  The lower water flowrates were sprayed
            through same number of nozzles, thus the spray
            drops were of larger size.
Inlet water flowrates:  0.76 I/sec in each section
oo
CO
Run
No.
41
42
4S
44
45
46
47
48
49
SO
Gas Inlet Conditions
Flow
C-BS!)1
Siun'
23.37
22.32
22.2
22.4
21.7
22.0
22.1
20.8
20.4
20.5
Temp
(°C)
11
45
45
46
56
57
56
66
68
66
Moisture
C* vol.)
.77
10.0
9.8
10.1
16.5
16.9
16.9
25. 6
26.7
26.8
Relative
Humidity
(t)
56
100
97
94
95
95
98
95
91
98
Liquid Temp.
(°C)
In
10.5
18.4
14.8
13.9
18.8
18.5
18.1
20.6
22.0
21.8
Out
10.1
26.4
12. S
22.7
31.3
31.3
30.8
40.0
42.5
41.3
Scrubber Pressure
Differences (cm w.C.)
&PS
0.8
0.7
0.7
0.7
0.8
0.8
0.8
0.8
0.8
0.8
iPE
-1.0
-0.9
-0.9
-0.8
-0.8
-0.8
-0.8
-0.8
-0.8
-0.8
&P0
-0.2
-0.2
-0.2
-0.2
	
	
	
	
...
	
Run
No.
51
52
53
54
55
56
57
58
Gas Inlet Conditions
Flow
(j*j
^min'
23.7
23.7
23.8
23.7
24.6
24.3
21.9
22.1
Temp
C°C)
17
15
10
12
14
10
46
47
Moisture
(S vol.)
1.4
1.2
1.1
1.1
1.0
0.9
1.4
1.3
Relative
Humidity
C»)
64
66
82
73
65
79
94
92
Liquid Temp.
(°C)
In
9.6
10.4
9.6
10.4
10.4
10.4
17.2
16.9
Out
11.5
11.5
10.4
11.2
»11.3
10.8
26.7
26.7
Scrubber Pressure
Differences (cm w.C.D
aps
0.4
0.4
0.4
0.4
0.9
0.9
0.4
0.5
&PE
-0.9
-0.9
-0.8
-0.9
-1.0
-1.0
-0.8
-0.8
AP0
-0.5
-0.5
-0.4
-O.S
-0.1
-0.1
-0.4
-0.3
Run
No.
41
42
43
44
45
46
47
48
49
50
Particulatc
Load x 10
In
72.3
77.1
99.8
121.5
72.4
84.2
97.8
66.4
161.9
77.2
>(B/l)Nm')
Out
20.2
9.7
11.5
10.1
4.1
2.6 ..
5.2
2.8
4.8
3.3
dn*
In
1.5
.78
.94
	
1.05
1.1
	
-.92
	
.96
Out
.82
.64
.85
Filti
1.05
.9
. Filti
1.1
Filti
.96
%
In
2.9
2.0
1.8
r Ru
1.6
1.8
r Ru
1.6
r Ru
1.7
Out
l.S
1.7
1.6
, 	
1.6
1.7
l 	
2.2
l 	
1.7
"i
xlO"'
1.0
7.7
2.9
	
l.ll
1.75
..; 	
1.5
	
1.7
q'
vin

.48
.49
.52
.94
.98
.99
1.66
1.77
1.78
F
Jet
27.9
12.6
11.5
8.3
5.7
3.0
S.3
4.25
2.95
4.27
Run
No.
51
52
53
54
55
56
57
58
Particulate
Load x 10
In
91.3
60.0
81.3
57.2
72.8
77.3
75.8
102.9
'(E/DUm1)
Out
56.2
45.1
49.4
39.9
31.4
33.8
16.1
24
dpE
In
1 .06
	
	
.98
1 .01
1.03
	
	
Out
1.04
Filte
Filte
1.03
.96
.94
Filte
Filte
%
In
1.8
Run
Run
1.7
1.8
1.7
Run
Run
Out
1.6
	
	
1.6
1.6
1.6
	
	
ni
xlO''
2.0
	
	
1.3
1.4
1.5
	
	
q1
xlO


...
...
...
...
.47
.47
f
61.6
75.2
60.8
69.7
43.2
43.7
21.2
23. S

-------
                     Table   I .4 (continued)
                                                                                                      Table   I .4  (continued)
Run
No.
59
60
61
62
63
64
65
66 '
Gas Inlet Conditions
Flow
DNm?
lininj
22
21.9
21.7
21.4
21.5
21.3
21.5
20.8
Temp
co
47
47
57
58
57
58
57
66
Moisture
(4 vol.)
1.3
1.4
.56
.59
.61
.57
.60
.83
Relative
Humidity
(»)
92
95
90
88
93
90
95
91
Liquid Temp.
CO
In
17.2
16.6
15.6
16.7
16.9
16
•14.1
19.6
Out
25.5
21.9
28.8
30
30
28.8
28.1
41.9
Scrubber Pressure
Differences (cm W.C.)
aps
0.5
0.4
0.5
0.5
0.5
0.5
0.5
0.5
*
iPE
-0.8
-0.7
-0.7
-0.7
-0.8
-0.8
-0.8
-0.8
AP0
'0.3
-0.3
-0.2
-0.2
-0.3
-0.3
-0.3
-0.3
Run
No.
67
68
69
70
71
72
73
74
Gas Inlet Conditions
Flow
( DNml
vmin'
19.9
20.9
20.8
20.5
19.4
19:8
19.3
19.1
Temp
CO
68
66
67
67
73
73
74
73
Moisture
(« vol.)
.68
.84
.62
.62
1.2
.8
1.2
1.1
Relative
Humidity
(t)
86
95
92
96
94
96
92
96
Liquid Temp.
CO
In
18.9
18.9
18
18.3.
22
20.8
22.3
22.2
Out
44.4
42.5
40.6
39.4
49.4
49.4
50.5
49.3
Scrubber Pressure
Differences (cm W.C.)
4PS
0.5
0.5
0.5
0.5
0.6
0.5
0.6
0.6
iPE
-0.8
-0.8
-0.7
-0.7
-0.7
-0.7
-0.7
-0.8
AP0
-0.3
-0.3
-0.2
-0.2
-0.1
-0.2
-0.1
-0.2
CO
Run
No.
59
60
61
62
63
64
65
66
Particulate
Load x 10
In
83.2
66.6
86.7
86.6
108.6
93.8
64.3
94.4 ..
Hg/DMnr)
Out
25.7
14.6
21.3
14.9
21.9
10.1
7.9
9.3
d
In
1.07
1.1
1.14
	
1.15
1.11
	
1.15
Out
.91
.95
.92
Filte;
.94
:92
Filte:
.89
'"a •"
In
1.7
1.8
1.6
Run
1.7
1-7
Run
1.7
Out
1.5
1.6
1.7
	
1.6
1:5
—
1.5
ni
xlO-s
.98
1.6
1.0
	
1.4
1.5
	
1.4.
q1
*in
.49
.49
.92
.97
.99
1.0
1.1
1.5
¥
30.9
21.9
24.5
17.2
20.2
10.8
12.3
9.8
Run
No.
67
68
69
70
71
72
73
74
Particulate
Load x 10'(g/DMnJ)
In
82.7
111.2
109.1
115.5
99.3
122.6
129.2
117.9
Out
10.5
21.8
10.7
13.8
13
10.3
12.7
9.7
dD«
In
1.25
	
	
1.2-
1.18
1.13
	
	
Out
.91
°E
In | Out
1.7 1.5
Filter Run 	
Filter Run 	
.95
.92
.95
1.8 1.6
1.8 1.6
1.8 1.5
Filter Run 	
Filter Run 	
ni
xlO't
.78

	
1.4
1.4
2.0
	
	
q'
iin
1.5
1.7
1.8
1.9
2.4
2.5
2.6
2.7
r
12.7
19.6
9.8
12'
13.1
8.4
9.8
8.2

-------
I .5  HORIZONTAL SPRAY SCRUBBER; OPERATING CONDITIONS AND PERFORMANCE

     Dust used:  Titanium dioxide, -16 mesh
     Spray configuration:  Scrubber operated in three stage mode.  Cold water
                 sprayed into the first section.  The drained water of section 1
                 resprayed into section 2 and the drain of section 2 sprayed into
                 section 3.  Thus, the scrubber is operated in co-current scheme..
     Spray water flow rates:  First Section:  1.05 liters/sec 6 2.67 Kg/cm1
                             Second Section:  1.09 liters/sec 8 2.67 Kg/cm1
                              Third Section:  1.0  liters/sec e 2.85 Kg/cm1
                                                                                                                   (continued)
Run
No.
75
76
77
78
79
Gas Inlet Condition
F10K, i Temp
(-22! ) ! °C
24.1
24.4
24.4
24.4
24.7
17
21
23
Moisturel Rel.
! vol. jtllun.
0.13
0.88
2.7
25 2.7
18.5 0.5
1
42
60
92
95
22
Liauid •
1st Section
In
12.5
13
13.5
14.5
10
Out
12.5
IS
1J.5
14.5
10
remnerature
2nd Section
•!n
12.5
13
13.5
12
10.5
Out
11.5
12
12
12.5
9
3rd Section
In
13
13.5
14.5
15.3
10.5
Out
11.5
11.8
12
12.5
8
Scrubber Pressure
Differences (cm W.C.)
1PS
0.8

-------
Table   1.5 (continued)
Run
No.
89
90
91
92
93
94
•95
96
Gas Inlet Condition
Flow
,DNn3,
(5Tm 5
20.1
20.1
19.6
20.7
18.2
19
18.8
18.7
Temp
°C
66
65.5
67
67
72.5
72.5
72.5
73
Moisture
S vol.
25.6
24.2 '
27.2
26.4
35.7
35.7
36.4
33.8
Rel.
Hum.
94
94
95
92
100
95
100
91
Liauid.
1st Section
In
17.5
17.5
17.5
16.5
16.5
17
17
17
Out
42.5
42
43.5
41
53.5
50.5
50.5
54
remperature
'2nd Section
In
34
34.5
33
34.5
44.5
40.5
41
42.5
Out
45.5
43
46
40
51.5
48.5
50
51
3rd Section-
In
41
40.5
40.5
40
51.5
48.5
50
51
Out
45.5 '
42.5
46
40 "
55
53
54.5
55.5
^.Scrubber Pressure
Differences (cm W.C.)
, APS
0.8
0.8
0.8 '
• 0.8
0.8
0.'8
0.9
0.9
A.PE
0.7
0.8
0:7
0.8
0.7
0.'7
0.9
0.8
AP0
0.1
0
0.1
0
0.1
0.1
0
0.1

Run

89
90
91
92
93

95
96

LoadxlO1
In
173
162
158
120

186
168

Cg/DNa3).
Out
14.7
15.1
12.7
6.48

12.4
15
I
°PE- V
In
'1.2
	
1.1
1.3

1.1
1.2
articulate
mA
Out
0.97
- Filter -
1.05
0.95

0.94
0;9

°,
In
1.8
— Run 	
1.8
lV7

2
1.8


Out
1.5
- --'--' 	 •
l.'S
1.6

1.5 '
' 1.6


l
2.2
	
2.6
4.1

5.4
2.8

x"lO

1.35
1.35
1.55
1.53

2.32
2.07

Pt
9*
>b
8,5
9.4
8.
5.4

6.7
85
   Table  . I  ..'6   HORIZONTAL  SPRAY SCRUBBER; OPERATING CONDITIONS AND PERFORMANCE

   Dust used:   Titanium  dioxide, -16 mesh                    •
   Spray  configuration:   Scrubber operated in three stage mode.  Cold water sprayed into
               the  third section and the scrubber operated  in counter-current scheme.
   Spray  water  flow rates:   First section:  1.05 4/sec 6 2.67 kg/cm2
                           Second section:  1.09 t/sec @ 2.67 kg/cm2
                            Third section:  1.0  I/sec 8 2.85 kg/cm!
Run
No.
97
98
99
100
101
102
Gas Inlet Condition
Flow
,DNm3,
lmTn '
23.4
24.0
23.4
23.2
22.8
23.2
Temp
"C
17.3
10.5
20
21
23
13
Moisture
* vol .
10.9
10.5
20
21
23
13
Rel.
tllun.
50
77
41
39
34
54
Liquid TemDerature
1st Section
In
7.7
10.7
9.9
9.2
9.1,
8.5
Out
12.0
10.4
11.2
12.0
10.7
9.3
2nd Section
In
7.2
9.9
9.9
9.1
9.9
8.9
Out
11.2
10.1
9.9
111?
9.7
9.3

3rd Section
In
10.1
12.5
11.7
12.3
13.2
12. -3
Out
11.3
10.4
10.9
1'2.8
11.5
9.6
Scrubber Pressure
Differences, (cm W.C.)
APS
0.9
0.4
0.9
' 0.8"
0.8
0.8 .
&Pf.
-1.0
-1.0
-1.0
-1.0
-1.0
-1.0
APQ
-0.2
-0.6
-0.1
-0.2
-0.2
-0.2

Run



99
101
102

LoadxlO3
In


I'OO'. 6
95.3'
79.4

(g/I'Nm3)
Out


44.5
41.4
36.0
E
V- V
In


0.98
0.96
0.96
articulate
mA
Out


0.92
0.9
0.88

°.
In


1.8
' 1.7
1.7


Out


1.5
1.5
1.5





2.6
2.0
1.7

q'



...
""-

Pt
9^
Sn

61 5
44.2
43.4
45.3
                                        91

-------
 Table   I.6 (continued)
Run
No.
103
104
105
106
107
108
109
110
111
Gas
Flow
rDNm',
'•mTn '
23.8
22.3
22.8
22.9
22.7
22.5
21.7
21.5
21.4
Inlet Condition
Temp
°C
11
46
46
46.5
46.5
47.5
56.5
56-
56.5
Moisture
\ vol.
1.1
10.0
10.0
10.0
10. 3
10.2
16.0
16.4
16.8
Rel.
Hum.
77
94
94
92
95
89
• 90
95
95
Liquid Tenoerature
1st Section
In
18.1
24.5
25.1
. 23.5
24.5
25.9
30.5
31.3
30.6
Out
12.8
31.5
29.8
31.5
30.0
30.5
43.8
44.4
43.8
2nd Section
In
17.3 .
.22.9
23.5
20.8
21.9
20.8
26.3
27.5
27.5
Out
12.8
29.0
26.0
26.7
26.5
26.5
37.5
38.8
40.0
3rd Section
In
19.7
18.7
17.1
14.4
18.1
13.3
14.4
16.7
17.3
Out
12.5
26.7
24.3
24.0
24.3
22.9
30
31.3
32.5
Scrubber Pressure
Differences (cm W.C.)
. 'PS
0.4
0.8
0.9
0.9
0.9
0.9
0.8
0.9
0.9
iPE
-0.9
-1.0
-1.1
-1.0
-1.0
-1.0
-0.9
-0.9
-1.0
iP0
-0.5
-0.2
-0.2
-0.1
-0.1
-0.1
-0.1
0.0
-0.1

Run
No

103
104
105

108
109
111


LoadxlO5
In
76.4
•106.3
68.4

68.3
96.6
177 <;
133 4


(g/D.Mn')
Out
49.8
29.8
16.0

11.5
13.6


F
flpg' V
In
0.96
0.94 .
1.0

0.94
1.1


articulate
mA
Out
1.0
0.84
0.82

0.92
0.82



a
G
In
1.8
1.6
1.6

1.7
1.6




Out
1.6
1.5
1.5

1.6
1.6





2.1
. 2-4
1.2

1.7
1.2



q'

0.47
0.47

0.52
0.87



Pt
%
65.2
28.0
23.4

16.8
14.1


Table   J.6  (continued)
Run
No.
112
113
114
115
116
117
118
119
120
Gas Inlet Condition
Flow
,DNm3,
(ijrrn- )
21.1
21.5
20.4
20.8
20.1
19.9
20.2
19.3
19.6
Temp
•c
56.5
57.5
66
66
66
68
67.5
74-
73
Moisture
t vol.
16.9
17.2
26.2
25.5 .
25.6
27.3
26.7
34.7
36.3
Rel.
Hun.
95
93
96'
93
93
91
91
90
100
Liquid Tercoerature
1st Section
In
31.9
33.1
43. 8«
43.8
47.0
45.6
45.0
52.3
50.5
Out
41.3
42.5
57.2
56.0
54.8
58.4
55.4
67.0
65.7
2nd Section
In
30.0
30.0
38.8
38.8
40.0
42.5
40.0
4 8. '8
48.8
Out
36.3
38.8
53.6
51.1
51.1
54.8
51.1
62.1
63.3
3rd Section
In
20.0
18.7
18.7
19.3
19.7
20.0
18.7
20.7
24.0
Out
35.6
30.6
42.5
41.3
40.0
43.8
41.3
52.3
52.3
Scrubber Pressure
Differences (cm IV.C.)
4PS
0.9
0.9
0.9
0.9
0.9
0.9
0.9
0.9
0.9
APE
-0.9
-1.0
-0.9
-0.9
-0.9
-0.9
-0.9
-0.9
-0.9
AP0
-0.1
0.0
-0.1
0.0
.0.0
.0.0
•0.0
0.0
0.0

Run

112
113
114
115
116
117
118
119
120


LoadxlO'
In
109.6
86.8
154.1
150.5
167.6
131.0
129 4
169 6


(g/DNm3)
Out
19.9
13.2
32.6
23.5
.42.0
24.1
24 3
20 0

I
V' V
In
1.03
1.2
1.03
1.1
	
1.0



articulate
mA
Out
0.85
0.82
0.82
0.88
	 Filter
0.86

Filter


a
In
1.5
1.6
1.7
1.6
Run 	
1.6





Out
1.5
1.5
1.6
1.5
	
1.5




« v lA^fi

1.3
0.7
2.5
2.1
	
2.6




q' .

0.9
0.95
1.5
1.5
1.5
1.5




Pt
*
18.2
15.2
21.2
15.6
25.1
18.4



                                        92

-------
Table   |  .6 (continued)
Run
No.
121
122
123
Gas Inlet Condition
Flow
,DNm?,
l'niin '
18.1
19.4
19.1
Temp
°C
74
73.5
71
Moisture
* vol.
36.3
36.3
33.2
Rel.
Hum.
94
96
98
Liauid Temperature
1st Section
In
47.5
48.1
49.9
Out
64.5
64.5
63.3
2nd Section
In
43.8
43'. 8
45.6
Out
60.9
60.9
59.7
3rd Section
In
21.3.
21.3
16.0
Out
50.0
49:9'
48.8
Scrubber Pressure
Differences (cm Iv'.C.J
APg
6.9
0.9
0.8
4PE
-0.8
-0.9
-0.8
AP0
0.1
0.0
0.0
Run
No.
121
122
123

Particulate
LoadxlO' (g/DNm3)
. In
135.4
98.3
76.8

Out
9.0
12.4
10.0

apg> pmA
In
1.23
1.12
1.05

Out
0.88
0.82
0.8
'
a
In
1.8
1.7
1.7

Out
1.6
1.5
1.4

n.^ x 10's
1.7 .
1.4
1.2

x 10
2.5
2.5
3.0
i
Pt
6.7
12.6
13.0

                                      93

-------
               0.5
          $    0.3
               0.2
              0.1
                                     RUN 12
                       RUN II
                      Cold  Run
                 0.3     0.5
                            p.
                                           2.0   3.0
                                                               0.3
                                                               0.2
                                                               0.1
                                                              O.OS
                                                              0'. 02
                                                                                       RUN 14
            g'  =  0.05
                                                                  0.2
                                                                              0.5
                                                                                        1.0
                                                                                V
                                                                                                   2.0   3.0
           Figure   1.1  - Particle penetration versus  aero-
                         dynamic diameter, single stage
                         spray scrubber.
     •Figure   | ,2 - Particle penetration versus aero-
                   dynanic diameter, single stage
                   spray scrubber.
     0.2
     0.1
    0.05
S
i
    0.01

      0.2
                g'  =  o.i
                — =  2.6H/m  /  section
                G	i      .    i   ,  i
                   0.5        1.0

                          dpa. uraA
                                         2.0   3.0  4.0
    Figure  ) .3. • Particle penetration versus aerodynamic
                  diameter, single stage spray scrubber.
                                                             0.2
                                                             0.1
                                                            O.OS
                                                            0.01
                                                                                           —  0.16
                                                                                                 0.16
                                                                   -  =  2.7   i/m3/sec.
                                                                0.2
                 0.5

                   d,
                                                                                       1.0
                                                                                                 2.0    3.0

Figure   I .4. - Particle penetration versus  aerodynamic
              diameter, single stage spray scrubber.
                                                 94

-------
   0.2
y...
So.05
  0.01
         —  =  3.4   £/m3  per,  sec.
     0.2
                   0.5


                    dpa>
                                        2.0    3.0
                                                             1.0
                                                             0.5
                                                         rt   o.i
                                                            0.05
                                                            0.03
                                                                                       »2i  g ' = 0 . 25 -
                                                                    G
                                                              0.2
                                                                          . 0.5
                                                                                       1.0
                                                                                                  2.0   3.0
  Figure   'I .5. - Particle  penetration versus aerodynamic
                 diameter, single stage spray scrubber.        Figure   |  .6. - Particle penetration versus aerodynamic
                                                                    	   diameter, single stage  spray scrubber
      1.0
      0.5
    o

    I
       0.1
                                      RUN »28  .
          0.3      0.5
                             1.0
                                              3.0    .«
                          dpa,
Figure   | .7.
                Particle penetration versus  aerodynamic
                diameter, single stage spray scrubber.
                                                            0.1
                                                           0.05
                                                          0.02
                                                                                          = 2 .6/sec  -
                                                               0.2
                                                                             0.5
                                                                                        1.0
                                                                                 dpa, umA
                                                                                                        3.0
                                                       Figure   1.8.
                                                                     Particle penetration  versus aerodynamic
                                                                     diameter, single stage spray scrubber.
                                                 95

-------
     0.3
     0.1
2   O.OB
    o.oi
                                          2.7/
                                          sec
                                          = 0.1  -,
         0.2
                       0.5
                                  1.0
                                                   3.0
                                                              1.1
                                                          £:    o.i
I  0.05
                                                              0.01
                                                                 0.2
                                                                                0.5
                                                                                            1.0
                                                                                                             3.0
                        V um
                                                                                     a.  yraA
    Figure   |  .9. - Particle penetration versus aerodynamic    Figure   ,  10  . Particle penetration versus aerodynamic
                   diameter, single  staje spray scrubber.                      diameter/single  stage spray scrubber.
     0.3
     0.2
  "•  0.1

  z
    0.05.
    0.02
         0.2
                                  l-.O
                                             2.0    3.0
   Figure   I  .11.  - Particle penetration  versus aerodynamic
                   diameter, single stage  spray scrubber.
                                                                 0.4
                                                            i
                                                                 0.1
                                                                 0.04
                                                                     0.3
                                                                            0.5
                                                                                        1.0

                                                                                        ymA
                                                                                                         3.0
                                                               Figure   | .12.  - Particle penetration versus aerodynamic
                                                                               diameter, tingle stage  spray scrubber.
                                                96

-------
  0.3
  0.2
z
o
f-
u
2 o.i
z
o
KH
i
  0.05
  0.01
                          • Run  042

                             q' =0.048
      0.3
  Figure   | .13
                          1.0
                                     2.0   3.0
                  -Particle  penetration versus^
                   aerodynamic  diameter, one
                   stage.
                                                    0.2
                                                  £0.1
                                                  o

                                                  i
                                                  o 0.05
                                                  u
                                                 .H 0.02
                                                   0.01
                                                                                   043
                                                                                q'=0.04&
                                                       0.3
                                                                           1.0
                                                                                      2.0    3.0
                                                   Figure   |,.14 - Particle penetration  versus
                                                                   aerodynamic diameter, one
                                                                   stage.
   0.1
 z
 o
   0.05
 u
 z
   0.01
                           q'=0.18
                i   iii
       0.3
                          1.0
                                     2.0     3.0
    Figure   |  .15 - Particle penetration versus
                   aerodynamic diameter, one
                   stage.
                                                 z
                                                 UJ
                                                 o.
                                                  D.
                                                      0.06

                                                      0.05

                                                      0.04


                                                      0.03



                                                      0.02
                                                      0.01
                                                     0.005
                                                          0.3     0.5
                                                                             1.0
                                                                                         2.0    3.0
                                                                         dpa,
                                                      Figure   |  .16 - Particle penetration versus
                                                                     aerodynamic diameter, single
                                                                     stage.
                                        97

-------
           cold  runs,  g "•  0
                                                    o
                                                    t-H
                                                    b
                                                         1.0
0.5
                                                    •?,    0.2
                                                    a.   0.1
                                                                   Run »55
                                                                  per  sec
   f  COld  runs,
         g'-   0
                                                                                  Run 56
                                                           0.3
                                                                             1.0   .     2.0    3.0
     Figure   1.17 - Particle penetration versus
                    aerodynamic diameter, one
                    stage.
Figure   | .18 -  Particle penetration versus
               aerodynamic diameter, one
               stage.

i
a.
    l.C
    0.5
    0.2
    0.1
                      Run  *59
         -L
         0.3
                          1.0
                                                         i.o.
                                     2.0    3.0
                     dpa,
    Figure  I  .19 - Particle penetration versus
                   aerodynamic diameter, one
                   stage.
                                                         0.5
                                                         0.-2
                                                         0.1
                                                        0»0«
                                                                      . Run  »61

                                                                           g.'   =• 0.091
       §-'•'
          per  sec
                                                            0.3
                                                                              1...0 .
                                                                                        2.0    3.0
              dpa, umA
                                                       Figure   | .20 - Particle penetration  versus
                                                                      aerodynamic diameter, one stage.
                                           98

-------
    0.4
    0.2
    0.1
*  0.05
  . 0.02
        0.3
                   per  sec
                          1.0

                        dpa, umA
                                     2.0     3.0
   Figure   I .21 -  Particle penetration versus
                   aerodynamic diameter, one  stage.
                                                         0.4


                                                         0.3
                                                    2    0.2
                                                    a:
'3
'u

 JZ
 a.
                                                         0.1
                                                        o.os
                                                        0.03
      Run »71

q'  =   0.24"


2.3  l/m3
per  sec
                                                            0.3
                                                                              1.0
                                                                                         2.0    3.0
                       pa. um


    Figure   I .22- Particle penetration versus
                  aerodynamic diameter, one stage.
    0.3
    0.2
    0.1
   0.05
                                                          1.0 |
   0.02
                                                          0.1
                                                                            RUN »7S
                                                             0.3     0.5
                                                                                i.o
                                                                         dpa, umA
                                                                                           RUN »79
                                                                                                 3.0
        0.3     '          1.0        2.0    3.0
                     dpa, gmA
   Figure  |  .23  -  Particle penetration  versus
                   aerodynamic diameter,  one stage.
                                                  Figure   1 .24 -  Particle penetration versus  aerodynamic
                                                                  diameter, 3 stage co-current spray scrubber.
                                               99

-------
  0.6
  0.06
      -  L
 =   0.045

-•  2.6  l/m3
   per  sec
      0.2
                              1.0
                                              3.0
                     pa,
 Figure   f. .25 -  Particle penetration versus aerodynamic
               diameter, 3 stage  co-current spray  scrubber.
                                    =0.135
                    0.5        1.0

                      dpa, UmA
                                               3.0
Figure   | .27 - Particle penetration  versus aerodynamic
              diameter, three stage co-current spray
              scrubber.
                                                              0.3
                                                              0.05
                                                                      RUN 186
                                                                                                 0.08?
                                                                           per   sec
                                                                 0.2
                                                                                dpa,  umA
                                                Figure   I .26 - Particle penetration versus aerodynamic
                                                              diameter, three  stage co-current spray
                                                              scrubber.
                                                              0.05
                                                                            q'=0.207
                                                                              0.5
                                                                                                          3.0
                                                                               dpa,  umA
                                               Figure   | .28  - Particle penetration versus aerodynamic
                                                    .  .       diameter, three stage co-current spray
                                                             scrubber.
                                                100

-------
     1.0
 i

  (-,  U.^
  Ul


  1  0.3
     0.2
     0.1
        0.3
                            1.0
                       dpa,  umA
                                      2.0    3.0
                                               §
                                                     z
                                                     o
                                                     z
                                                     la
                                                     a.
                                                         1.0
                                                   o.s

                                                   0.4

                                                   0.3
                                                         0.2
                                                         0.1
                                                                                Run »103
                                                            .03
                                                                   per  sec
                                                                         1.0

                                                                   dpa, umA
                                                                                          2.0    3.0
     Figure   I.29 - Particle penetration versus
                     aerodynamic diameter, three
                     stage counter-current.
                                                   Figure    | .30  - Particle penetration versus
                                                                  aerodynamic diameter, three
                                                                  stage counter-current.
§
z
o
uj
*j
u
0.1


0.3



0.2
      0.1
      0.05
          0.3
      Figure
                               Run 1104
g'
L
G
 Run I10S




 =  0.047

=  2.6  l/m*
                            i.o
                                       2.0
                                              3.0
        |  .31,- Particle penetration versus
               aerodynamic diameter, three
               stage counter-current.
                                                    z
                                                    o
                                              z
                                              o

                                             I
                                              UJ
                                              z
                                                         0.3
                                                         0.2 _
                                                         0.01
                                                            0.3
                                                                               1.0
                                                                                          2.0    3.0
                                                                         dpa,
                                                        Figure   I  .32 - Particle penetration versus
                                                                       aerodynamic diameter,  three
                                                                       stage counter current.
                                             101

-------
      8.7
§
      0.3
      0.2
      0.1
     O.OS


     0.04


     0.03
                                =  0.09   -
                           i
                                                     i
                                                     114
 0.4


 0.3



 0.2







 0.1






0.05
                                                          0.02
                                                        g'   =  0.25

                                                         T3'1
                                                         G   H/m3   .
          0.3
                            1.0
                                       2.0   3.0
                                                                0.3
                                                                                  1.0
                                                                                             2.0    3.0
                        upa'  »"""


     Figure   I  .33 - Particle  penetration versus
                    aerodynamic diameter, three
                    stage counter-current.
                                                  •  "pa- "'""

                                Figure   I .34  -  Particle penetration  versus
                                                aerodynamic diameter, three
                                                stage .counter-current.
                               0.6,



                               0.4


                               0.3



                               0.2
                               0.1
<   0.05
CL,
                              0.03
                                                      Run »117
                                  0.3
                                                    1.0
                                                               2.0    3.0
                                                 pg,
                             Figure  | .35- Particle penetration versus
                                           aerodynamic diameter, three
                                           stage counter-current
                                            102

-------
                                 TECHNICAL REPORT DATA
                          (Please read Instructions on the reverse before completing)
 1. REPORT NO.
  EPA-600/2-76-200
                            2.
             3. RECIPIENT'S ACCESSION NO.
 4. TITLE AND SUBTITLE
 Study of Horizontal-Spray Flux Force/Condensation
  Scrubber
            5. REPORT DATE
             July 1976
            6. PERFORMING ORGANIZATION CODE
 7. AUTHOR(S)
                                                      8. PERFORMING ORGANIZATION REPORT NO
 Seymour Calvert and Shui-Chow Yung
 9. PERFORMING ORSANIZATION NAME AND ADDRESS
 A. P. T. , Inc.
 4901 Morena Boulevard (Suite 402)
 San Diego, CA  92117
             10. PROGRAM ELEMENT NO.
             1AB012; ROAP 21ADL-002
             11. CONTRACT/GRANT NO.

             68-02-1328,  Task 10
 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
             Task Final; 2-12/75	
            14. SPONSORING AGENCY CODE
             EPA-ORD
 15.SUPPLEMENTARY NOTESEPA tagk officer for this report is L.E.  Sparks, Mail Drop 61,
 Ext 2925.
 16. ABSTRACT
              report gives results of B. laboratory pilot-scale evaluation of a Flux
 Force/Condensation (FF/C) scrubber for collecting fine particles, those smaller
 than 2 micrometers in diameter.  FF/C scrubbing includes the effects of diffusio-
 phoresis, thermophoresis , Stefan flow,  and particle growth due to water vapor con-
 densation.  The FF/C scrubber tested was of horizontal spray configurations. Effects
 of the scrubber configurations , liquid and gas flowrates ,  particle number concentra-
 tion, and the amount of vapor condensation were studied experimentally.  Fractional
 particle penetrations were measured with cascade impactors  and  are presented in
 terms of particle penetration as a function of particle size. The experimental results
 are compared with predictions from mathematical models. Optimum operational
 regions  and the technical and economic feasibility of FF/C scrubbing are determined
 and demonstrated for a fine particle pollution source.  It was confirmed that FF/C
 scrubbing is an attractive control method for fine particles when high efficiency is
 required or when the gas is hot enough to evaporate the necessary water vapor for
 condensation in the scrubber.  A program to demonstrate FF/C scrubbing at pilot
 scale for the control of fine particulate emissions from industrial sources is des-
 cribed.
 7.
                             KEY WORDS AND DOCUMENT ANALYSIS
                DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS  C.  COSATI Field/Group
 Air Pollution
 Scrubbers
 Condensing
 Dust
Air Pollution Control
Stationary Sources
Flux Force/Conden-
 sation
Particulate
13B
07A
07D
11G
 8. DISTRIBUTION STATEMENT

 Unlimited
19. SECURITY CLASS (ThisReport)
Unclassified
21. NO. OF PAGES

    117
20. SECURITY CLASS (Thispage)
Unclassified
                         22. PRICE
EPA Form 2220-1 (9-73)
                                          103

-------